{ "$schema": "https://themachinepress.com/schemas/edition-v1.schema.json", "schema_version": "1.0.0", "document_type": "machine_press_edition", "edition_id": "mp-2026-11-02-morning-0115", "source_edition_id": "mp-2026-11-02-morning-0115", "edition_number": 115, "edition_label": "Morning edition", "status": "published", "revision": 1, "published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "timezone": "America/Detroit", "canonical_url": "https://themachinepress.com/edition/2026-11-02", "html_url": "https://themachinepress.com/edition/2026-11-02", "json_url": "https://themachinepress.com/edition/2026-11-02.json", "markdown_url": "https://themachinepress.com/edition/2026-11-02.md", "lead_story_id": "mp-2026-11-02-001", "lead_summary": "Every multi-reader panel improved chest X-ray severity prediction over the no-gaze baseline, while no individual reader reached significance on the main endpoint.", "coverage_window": { "start": "2026-10-30T12:47:34.000-04:00", "end": "2026-11-02T08:25:20.000-05:00" }, "story_count": 27, "stories": [ { "story_id": "mp-2026-11-02-001", "source_story_id": "tmp-lead-gazerib-reader-panels", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 1, "story_type": "lead", "section": "research", "editorial_classification": "editorial", "headline": "Three Readers Capture Most of the Gaze Signal", "slug": "three-readers-capture-most-of-the-gaze-signal", "dek": "Every multi-reader panel improved chest X-ray severity prediction over the no-gaze baseline, while no individual reader reached significance on the main endpoint.", "summary": "Every multi-reader panel improved chest X-ray severity prediction over the no-gaze baseline, while no individual reader reached significance on the main endpoint.", "body_text": "GazeRIB asks whether radiologists' visual attention can supervise a model more reliably when it is pooled across readers. Eight experts repeatedly read the same 80 chest X-rays; the study compared individual readers with three-, five- and eight-reader panels across seven architectures and ten seeds, then evaluated 910 models on a separate 2,897-image cohort. Every panel improved macro F1 by 5.33 to 7.47 percentage points over the no-gaze baseline, while no individual reader reached significance on that endpoint. Three readers captured 73 to 80 percent of the eight-reader gain across macro F1, ROC-AUC and precision-recall area. External transfer was mixed: panels improved ROC-AUC on SIIM-FISABIO-RSNA, but not on MIMIC-Eye. The result supports panel composition as an experimental variable, not a claim that gaze guidance will transfer to every clinical setting.", "why_it_matters": "Every multi-reader panel improved chest X-ray severity prediction over the no-gaze baseline, while no individual reader reached significance on the main endpoint.", "limitations": [ "External transfer was mixed: panels improved ROC-AUC on SIIM-FISABIO-RSNA, but not on MIMIC-Eye.", "The result supports panel composition as an experimental variable, not a claim that gaze guidance will transfer to every clinical setting." ], "importance": 10, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-001/three-readers-capture-most-of-the-gaze-signal", "json_url": "https://themachinepress.com/story/mp-2026-11-02-001.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-001-001", "text": "Every multi-reader panel improved chest X-ray severity prediction over the no-gaze baseline, while no individual reader reached significance on the main endpoint.", "source_ids": [ "source-2026-11-02-001" ], "qualification": "External transfer was mixed: panels improved ROC-AUC on SIIM-FISABIO-RSNA, but not on MIMIC-Eye." } ], "source_ids": [ "source-2026-11-02-001" ], "tags": [ "medical imaging", "radiology", "gaze supervision", "evaluation" ], "image_url": "https://themachinepress.com/issues/2026-11-02/lead-radiologist-gaze.png", "corrections": [] }, { "story_id": "mp-2026-11-02-002", "source_story_id": "tmp-feature-roboarchitect-harness", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 2, "story_type": "secondary", "section": "robotics", "editorial_classification": "editorial", "headline": "The Policy Stays Frozen. The Harness Learns.", "slug": "the-policy-stays-frozen-the-harness-learns", "dek": "RoboArchitect let a coding model revise the control harness around a frozen robot policy, raising reported success from 9.7 to 43.3 percent on unseen scenes.", "summary": "RoboArchitect let a coding model revise the control harness around a frozen robot policy, raising reported success from 9.7 to 43.3 percent on unseen scenes.", "body_text": "A robot policy does not act alone: surrounding control code decides when to query it, call a vision tool or command the arm directly. RoboArchitect automates the design of that harness. A coding model watches rollouts, diagnoses failures and iteratively revises the controller around an unchanged policy; evaluation then moves to held-out scenes and tasks. Across the reported benchmarks and policies, average success rose from 9.7 to 43.3 percent on unseen scenes and from 12.7 to 40.2 percent on unseen tasks. The evolved harnesses discovered tactics such as lifting a jammed arm clear, approaching around obstacles and retaining useful grasp depths in persistent memory. The evidence is benchmark-based, and it does not establish that unconstrained self-modification is safe or robust on arbitrary hardware.", "why_it_matters": "RoboArchitect let a coding model revise the control harness around a frozen robot policy, raising reported success from 9.7 to 43.3 percent on unseen scenes.", "limitations": [ "A robot policy does not act alone: surrounding control code decides when to query it, call a vision tool or command the arm directly.", "The evidence is benchmark-based, and it does not establish that unconstrained self-modification is safe or robust on arbitrary hardware." ], "importance": 10, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-002/the-policy-stays-frozen-the-harness-learns", "json_url": "https://themachinepress.com/story/mp-2026-11-02-002.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-002-001", "text": "RoboArchitect let a coding model revise the control harness around a frozen robot policy, raising reported success from 9.7 to 43.3 percent on unseen scenes.", "source_ids": [ "source-2026-11-02-002" ], "qualification": "A robot policy does not act alone: surrounding control code decides when to query it, call a vision tool or command the arm directly." } ], "source_ids": [ "source-2026-11-02-002" ], "tags": [ "robotics", "agent harnesses", "persistent memory", "policy adaptation" ], "image_url": "https://themachinepress.com/issues/2026-11-02/feature-roboarchitect.png", "corrections": [] }, { "story_id": "mp-2026-11-02-003", "source_story_id": "tmp-story-larc-latent-reasoning", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 3, "story_type": "dispatch", "section": "media-creative-tools", "editorial_classification": "editorial", "headline": "Image Reasoning Moves Into the Latent Space", "slug": "image-reasoning-moves-into-the-latent-space", "dek": "LARC replaces token-level image reasoning with continuous hidden-state reasoning and reports a seven-point GenEval gain over BAGEL.", "summary": "LARC replaces token-level image reasoning with continuous hidden-state reasoning and reports a seven-point GenEval gain over BAGEL.", "body_text": "Unified multimodal models often reason in discrete language tokens before generating in a continuous image space. LARC treats that modality mismatch as the bottleneck. Its curriculum first converts token reasoning into continuous latent reasoning, then an information-gain reinforcement-learning stage continues without ground-truth reasoning traces. Across the reported text-to-image evaluations, the method improved generation quality, prompt following and compositional fidelity; on GenEval it beat BAGEL by seven percentage points and showed gains from additional test-time computation. Those are benchmark results from the authors, not evidence that the hidden reasoning is human-readable or causally faithful.", "why_it_matters": "LARC replaces token-level image reasoning with continuous hidden-state reasoning and reports a seven-point GenEval gain over BAGEL.", "limitations": [ "Those are benchmark results from the authors, not evidence that the hidden reasoning is human-readable or causally faithful." ], "importance": 9, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-003/image-reasoning-moves-into-the-latent-space", "json_url": "https://themachinepress.com/story/mp-2026-11-02-003.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-003-001", "text": "LARC replaces token-level image reasoning with continuous hidden-state reasoning and reports a seven-point GenEval gain over BAGEL.", "source_ids": [ "source-2026-11-02-003" ], "qualification": "Those are benchmark results from the authors, not evidence that the hidden reasoning is human-readable or causally faithful." } ], "source_ids": [ "source-2026-11-02-003" ], "tags": [ "multimodal models", "image generation", "latent reasoning", "test-time scaling" ], "image_url": "https://themachinepress.com/issues/2026-11-02/latent-reasoning.webp", "corrections": [] }, { "story_id": "mp-2026-11-02-004", "source_story_id": "tmp-story-flide-exact-forking", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 4, "story_type": "dispatch", "section": "infrastructure", "editorial_classification": "editorial", "headline": "Decision Trees Stop Recomputing Their Trunks", "slug": "decision-trees-stop-recomputing-their-trunks", "dek": "Flide flattens branching decision requests into one execution stream and reports answers as much as 9.5 times faster than vLLM and SGLang.", "summary": "Flide flattens branching decision requests into one execution stream and reports answers as much as 9.5 times faster than vLLM and SGLang.", "body_text": "Decision models repeatedly evaluate branches that share a long prefix, but generation-oriented engines either recompute that prefix or alter output probabilities. Flide separates the logical request tree from its physical execution. ForkAttention uses GPU-resident indices across standard attention and recurrent Gated DeltaNet layers so each branch is equivalent, in exact arithmetic, to an isolated pass. The engine also resumes expanding histories from earlier checkpoints. On the reported public models and workloads, Flide answered queries up to 9.5 times faster than vLLM and SGLang while staying closest to a float32 reference in total-variation distance. The speedup is workload-specific and does not imply the same gain for ordinary free-form generation.", "why_it_matters": "Flide flattens branching decision requests into one execution stream and reports answers as much as 9.5 times faster than vLLM and SGLang.", "limitations": [ "The speedup is workload-specific and does not imply the same gain for ordinary free-form generation." ], "importance": 8, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-004/decision-trees-stop-recomputing-their-trunks", "json_url": "https://themachinepress.com/story/mp-2026-11-02-004.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-004-001", "text": "Flide flattens branching decision requests into one execution stream and reports answers as much as 9.5 times faster than vLLM and SGLang.", "source_ids": [ "source-2026-11-02-004" ], "qualification": "The speedup is workload-specific and does not imply the same gain for ordinary free-form generation." } ], "source_ids": [ "source-2026-11-02-004" ], "tags": [ "inference", "decision models", "GPU systems", "agents" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-005", "source_story_id": "tmp-story-latent-sparse-attention", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 5, "story_type": "dispatch", "section": "frontier-models", "editorial_classification": "editorial", "headline": "Ten Million Tokens Get a Hierarchical Index", "slug": "ten-million-tokens-get-a-hierarchical-index", "dek": "Latent Sparse Attention clusters keys in a learned hierarchy and reports a 72.4-times layer-level prefill speedup over full attention at ten million tokens.", "summary": "Latent Sparse Attention clusters keys in a learned hierarchy and reports a 72.4-times layer-level prefill speedup over full attention at ten million tokens.", "body_text": "Sparse attention still struggles when its indexer either touches every key or loses accuracy at extreme context lengths. Latent Sparse Attention clusters keys hierarchically in a learned latent space and lets each query score clusters instead of all keys. A distilled version can retrofit pretrained models; a native version learns from next-token loss. On the reported GLM-5.3-FLASH test at ten million tokens, one layer ran 22.7 times faster in prefill than the model's original sparse attention and 72.4 times faster than full attention, with similar task accuracy; decode gains were 4.6 and 42.5 times respectively. These are layer-level results, not end-to-end application latency guarantees.", "why_it_matters": "Latent Sparse Attention clusters keys in a learned hierarchy and reports a 72.4-times layer-level prefill speedup over full attention at ten million tokens.", "limitations": [ "These are layer-level results, not end-to-end application latency guarantees." ], "importance": 8, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-005/ten-million-tokens-get-a-hierarchical-index", "json_url": "https://themachinepress.com/story/mp-2026-11-02-005.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-005-001", "text": "Latent Sparse Attention clusters keys in a learned hierarchy and reports a 72.4-times layer-level prefill speedup over full attention at ten million tokens.", "source_ids": [ "source-2026-11-02-005" ], "qualification": "These are layer-level results, not end-to-end application latency guarantees." } ], "source_ids": [ "source-2026-11-02-005" ], "tags": [ "long context", "sparse attention", "indexing", "efficiency" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-006", "source_story_id": "tmp-story-idwm-implicit-dynamics", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 6, "story_type": "dispatch", "section": "robotics", "editorial_classification": "editorial", "headline": "The Planner Chooses Reachable States Before Exact Actions", "slug": "the-planner-chooses-reachable-states-before-exact-actions", "dek": "ID-WM separates high-level reachability from low-level inverse dynamics and reports a 135 percent gain over single-level world-model control.", "summary": "ID-WM separates high-level reachability from low-level inverse dynamics and reports a 135 percent gain over single-level world-model control.", "body_text": "Long-horizon world-model planning accumulates errors when one model must choose distant goals and exact actions at once. ID-WM divides the problem: a high-level model samples feasible state transitions, while a low-level world model solves the inverse dynamics needed to realize each transition. That lets the system assemble new plans from short behavioral segments without committing early to rigid macro-actions. Across three navigation and manipulation suites, the authors report a 38.5 percent improvement over policy and diffusion-based planners and a 135 percent gain over single-level world-model model-predictive control in complex unseen configurations. The scope is the tested suites; physical deployment and safety are not established by the abstract.", "why_it_matters": "ID-WM separates high-level reachability from low-level inverse dynamics and reports a 135 percent gain over single-level world-model control.", "limitations": [ "The scope is the tested suites; physical deployment and safety are not established by the abstract." ], "importance": 8, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-006/the-planner-chooses-reachable-states-before-exact-actions", "json_url": "https://themachinepress.com/story/mp-2026-11-02-006.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-006-001", "text": "ID-WM separates high-level reachability from low-level inverse dynamics and reports a 135 percent gain over single-level world-model control.", "source_ids": [ "source-2026-11-02-006" ], "qualification": "The scope is the tested suites; physical deployment and safety are not established by the abstract." } ], "source_ids": [ "source-2026-11-02-006" ], "tags": [ "world models", "planning", "inverse dynamics", "robotics" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-007", "source_story_id": "tmp-story-dittany-mas-repair", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 7, "story_type": "dispatch", "section": "developer-tools", "editorial_classification": "editorial", "headline": "Multi-Agent Repair Keeps the Work That Still Counts", "slug": "multi-agent-repair-keeps-the-work-that-still-counts", "dek": "Dittany preserves validated intermediate contributions while repairing unmet conditions, reaching 84.76 percent task accuracy across 9,000 tasks.", "summary": "Dittany preserves validated intermediate contributions while repairing unmet conditions, reaching 84.76 percent task accuracy across 9,000 tasks.", "body_text": "A multi-agent run can fail its final requirement even when several intermediate results remain useful. Dittany derives expected contributions and acceptance conditions from the task and method specifications, links them in a contribution graph, and checks which outputs still satisfy their role. It then attempts action-level correction when available inputs suffice or graph-level recovery when dependencies must change. Across 30 applications, six frameworks and 9,000 tasks, the paper reports 88.28 percent failure-detection precision, 79.18 percent recall and 84.76 percent task accuracy—26.74 points above native execution and 21.28 above the strongest tested recovery baseline. Deployment token use remained 17.92 percent above native execution.", "why_it_matters": "Dittany preserves validated intermediate contributions while repairing unmet conditions, reaching 84.76 percent task accuracy across 9,000 tasks.", "limitations": [ "A multi-agent run can fail its final requirement even when several intermediate results remain useful." ], "importance": 8, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-007/multi-agent-repair-keeps-the-work-that-still-counts", "json_url": "https://themachinepress.com/story/mp-2026-11-02-007.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-007-001", "text": "Dittany preserves validated intermediate contributions while repairing unmet conditions, reaching 84.76 percent task accuracy across 9,000 tasks.", "source_ids": [ "source-2026-11-02-007" ], "qualification": "A multi-agent run can fail its final requirement even when several intermediate results remain useful." } ], "source_ids": [ "source-2026-11-02-007" ], "tags": [ "multi-agent systems", "runtime repair", "contribution graphs", "reliability" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-008", "source_story_id": "tmp-story-private-observable-batches", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 8, "story_type": "dispatch", "section": "safety-security", "editorial_classification": "editorial", "headline": "Private Batches Hide Their Size With Dummy Points", "slug": "private-batches-hide-their-size-with-dummy-points", "dek": "A randomized dummy-point mechanism protects variable batch counts even when observers can see traffic and systems must truncate oversized batches.", "summary": "A randomized dummy-point mechanism protects variable batch counts even when observers can see traffic and systems must truncate oversized batches.", "body_text": "Privacy amplification for DP-SGD often assumes that random batch sizes stay secret. In distributed systems, an aggregator or traffic observer may see those counts, while memory limits may force oversized batches to be truncated—both conditions can weaken the guarantee. The proposed mechanism adds a random number of dummy points so the visible count is privatized while the real batch can still be truncated. The analysis covers Poisson sampling and random allocation and uses subsampling to reduce the expected dummy overhead below direct non-negative-noise protection of sensitivity-one counts. For truncated Poisson batches, the authors report a better privacy-efficiency tradeoff than recent direct accounting. The result is a mechanism and analysis, not an audit of a deployed DP system.", "why_it_matters": "A randomized dummy-point mechanism protects variable batch counts even when observers can see traffic and systems must truncate oversized batches.", "limitations": [ "The result is a mechanism and analysis, not an audit of a deployed DP system." ], "importance": 8, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-008/private-batches-hide-their-size-with-dummy-points", "json_url": "https://themachinepress.com/story/mp-2026-11-02-008.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-008-001", "text": "A randomized dummy-point mechanism protects variable batch counts even when observers can see traffic and systems must truncate oversized batches.", "source_ids": [ "source-2026-11-02-008" ], "qualification": "The result is a mechanism and analysis, not an audit of a deployed DP system." } ], "source_ids": [ "source-2026-11-02-008" ], "tags": [ "differential privacy", "DP-SGD", "distributed systems", "batching" ], "image_url": "https://themachinepress.com/issues/2026-11-02/private-batches.webp", "corrections": [] }, { "story_id": "mp-2026-11-02-009", "source_story_id": "tmp-story-relift-trajectory-repair", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 9, "story_type": "dispatch", "section": "research", "editorial_classification": "editorial", "headline": "The 3D Tracker Repairs the Lifted Trajectory", "slug": "the-3d-tracker-repairs-the-lifted-trajectory", "dek": "ReLift treats dense monocular 3D tracking as geometric correction and completion instead of relearning correspondence, depth and occlusion together.", "summary": "ReLift treats dense monocular 3D tracking as geometric correction and completion instead of relearning correspondence, depth and occlusion together.", "body_text": "Strong pretrained 2D trackers and geometry models can already lift video points into 3D, but their trajectories inherit correspondence mistakes, depth errors and occluder depth during gaps. ReLift focuses 4D supervision on repairing those trajectories. Its feed-forward corrector alternates attention along and across tracks, preserves reliable observations, fixes corrupted positions and completes missing motion. Because it trains on geometric trajectories with simulated lifting errors, it can work across upstream trackers and geometry providers without retraining. On WorldTrack, the paper reports state-of-the-art mean accuracy, improving the strongest baselines by 2.1 APD overall and 3.0 on dynamic points. Those gains depend on the reported benchmark and input components.", "why_it_matters": "ReLift treats dense monocular 3D tracking as geometric correction and completion instead of relearning correspondence, depth and occlusion together.", "limitations": [], "importance": 7, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-009/the-3d-tracker-repairs-the-lifted-trajectory", "json_url": "https://themachinepress.com/story/mp-2026-11-02-009.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-009-001", "text": "ReLift treats dense monocular 3D tracking as geometric correction and completion instead of relearning correspondence, depth and occlusion together.", "source_ids": [ "source-2026-11-02-009" ], "qualification": null } ], "source_ids": [ "source-2026-11-02-009" ], "tags": [ "3D tracking", "computer vision", "trajectory repair", "occlusion" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-010", "source_story_id": "tmp-story-ai-weather-ensemble-blend", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 10, "story_type": "dispatch", "section": "research", "editorial_classification": "editorial", "headline": "The Best Weather Blend Needs Different Kinds of Error", "slug": "the-best-weather-blend-needs-different-kinds-of-error", "dek": "Blending an AI ensemble with a physics-based ensemble improved UK temperature and wind forecasts most when the systems made complementary mistakes.", "summary": "Blending an AI ensemble with a physics-based ensemble improved UK temperature and wind forecasts most when the systems made complementary mistakes.", "body_text": "Forecast centers blend multiple ensemble systems because diversity can improve probabilistic skill. This study pairs the ensemble version of AIFS-CRPS with the Met Office's convection-permitting MOGREPS-UK and examines UK site temperature and wind speed. The combined forecasts improved overall skill across locations, lead times and exceedance thresholds, particularly when the component models had different error characteristics. Decomposing continuous ranked probability score showed that both model skill and diversity mattered: MOGREPS-UK paired better with either AIFS-CRPS or IFS than IFS paired with AIFS-CRPS, whose errors were less diverse. Calibration before and after blending improved accuracy further. The conclusion concerns the tested forecasts and region, not every AI-physics ensemble.", "why_it_matters": "Blending an AI ensemble with a physics-based ensemble improved UK temperature and wind forecasts most when the systems made complementary mistakes.", "limitations": [ "The conclusion concerns the tested forecasts and region, not every AI-physics ensemble." ], "importance": 8, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-010/the-best-weather-blend-needs-different-kinds-of-error", "json_url": "https://themachinepress.com/story/mp-2026-11-02-010.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-010-001", "text": "Blending an AI ensemble with a physics-based ensemble improved UK temperature and wind forecasts most when the systems made complementary mistakes.", "source_ids": [ "source-2026-11-02-010" ], "qualification": "The conclusion concerns the tested forecasts and region, not every AI-physics ensemble." } ], "source_ids": [ "source-2026-11-02-010" ], "tags": [ "weather forecasting", "ensembles", "AI models", "probabilistic prediction" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-011", "source_story_id": "tmp-story-ads-crash-composition", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 11, "story_type": "dispatch", "section": "safety", "editorial_classification": "editorial", "headline": "Crash Composition Offers a Mileage-Free Comparison", "slug": "crash-composition-offers-a-mileage-free-comparison", "dek": "A harmonized study compares the mix—not the rate—of automated-driving crashes with human baselines and finds two differences robust to its reporting assumptions.", "summary": "A harmonized study compares the mix—not the rate—of automated-driving crashes with human baselines and finds two differences robust to its reporting assumptions.", "body_text": "Most automated-driving operators do not publish a usable exposure denominator, so this study avoids crash-rate claims and asks whether the composition of crash types differs. It harmonizes 2,450 NHTSA Standing General Order incidents, with 828 inside the 2021–2024 human-comparator window, and uses a schema-constrained language model to map narratives into the federal CRSS taxonomy. Turn-across-path and single-vehicle crashes were under-represented among automated-driving crashes under every tested reportability, towing, weighting and driverless definition. The author also flags operator-specific driverless coding, fleet-recovery effects on towing and an occupancy confound. Because exposure is absent, the analysis cannot establish whether automated systems crash more or less often overall.", "why_it_matters": "A harmonized study compares the mix—not the rate—of automated-driving crashes with human baselines and finds two differences robust to its reporting assumptions.", "limitations": [ "Most automated-driving operators do not publish a usable exposure denominator, so this study avoids crash-rate claims and asks whether the composition of crash types differs.", "Because exposure is absent, the analysis cannot establish whether automated systems crash more or less often overall." ], "importance": 8, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-011/crash-composition-offers-a-mileage-free-comparison", "json_url": "https://themachinepress.com/story/mp-2026-11-02-011.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-011-001", "text": "A harmonized study compares the mix—not the rate—of automated-driving crashes with human baselines and finds two differences robust to its reporting assumptions.", "source_ids": [ "source-2026-11-02-011" ], "qualification": "Most automated-driving operators do not publish a usable exposure denominator, so this study avoids crash-rate claims and asks whether the composition of crash types differs." } ], "source_ids": [ "source-2026-11-02-011" ], "tags": [ "automated driving", "crash data", "NHTSA", "evaluation" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-012", "source_story_id": "tmp-story-graphene-haldane-gap", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 12, "story_type": "dispatch", "section": "research", "editorial_classification": "editorial", "headline": "Circular Light Opens Graphene's Haldane Gap", "slug": "circular-light-opens-graphene-s-haldane-gap", "dek": "Time-resolved photoemission directly observed a Dirac gap that appeared with circularly polarized driving and vanished with the optical field.", "summary": "Time-resolved photoemission directly observed a Dirac gap that appeared with circularly polarized driving and vanished with the optical field.", "body_text": "Circularly polarized light has long been predicted to break time-reversal symmetry in graphene and open a topological gap at the Dirac point. The reported time- and angle-resolved photoemission experiment observes that gap directly in monolayer graphene: it appears with the optical field and only under circular polarization. Paired with Floquet theory, the authors identify the feature with a light-induced Haldane mass and infer the associated Berry curvature and Chern topology. The experiment establishes an ultrafast route to dynamically altering symmetry and topology in the tested material and conditions; it does not yet amount to a general-purpose topological device.", "why_it_matters": "Time-resolved photoemission directly observed a Dirac gap that appeared with circularly polarized driving and vanished with the optical field.", "limitations": [ "The reported time- and angle-resolved photoemission experiment observes that gap directly in monolayer graphene: it appears with the optical field and only under circular polarization.", "The experiment establishes an ultrafast route to dynamically altering symmetry and topology in the tested material and conditions; it does not yet amount to a general-purpose topological device." ], "importance": 9, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-012/circular-light-opens-graphene-s-haldane-gap", "json_url": "https://themachinepress.com/story/mp-2026-11-02-012.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-012-001", "text": "Time-resolved photoemission directly observed a Dirac gap that appeared with circularly polarized driving and vanished with the optical field.", "source_ids": [ "source-2026-11-02-012" ], "qualification": "The reported time- and angle-resolved photoemission experiment observes that gap directly in monolayer graphene: it appears with the optical field and only under circular polarization." } ], "source_ids": [ "source-2026-11-02-012" ], "tags": [ "graphene", "topological matter", "Floquet engineering", "photoemission" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-013", "source_story_id": "tmp-story-roof-humanoid-recovery", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 13, "story_type": "dispatch", "section": "robotics", "editorial_classification": "editorial", "headline": "The Fallen Humanoid Learns to Make Room First", "slug": "the-fallen-humanoid-learns-to-make-room-first", "dek": "ROOF rewards a robot for reducing overlap with an overhead obstacle before standing, lifting simulated recovery from 18.9 to 88.5 percent in one setting.", "summary": "ROOF rewards a robot for reducing overlap with an overhead obstacle before standing, lifting simulated recovery from 18.9 to 88.5 percent in one setting.", "body_text": "A humanoid trapped beneath furniture cannot simply execute a standard stand-up motion; it must create clearance while building stable support. ROOF supplies contact-supported geometric guidance that rewards reducing robot-obstacle overlap without prescribing an escape path, then combines it with a learned motion prior and posture-aware support and joint-stall costs. In simulation beneath lift-only plates, recovery reached 88.5 percent versus 18.9 percent without the geometric guidance; beneath freely moving randomized plates it reached 98.6 percent. A Unitree G1 recovered in 95 of 150 attempts across seven real obstacle scenes. The real-world success rate also shows the remaining gap, and the policy was tested on one hardware platform.", "why_it_matters": "ROOF rewards a robot for reducing overlap with an overhead obstacle before standing, lifting simulated recovery from 18.9 to 88.5 percent in one setting.", "limitations": [ "A humanoid trapped beneath furniture cannot simply execute a standard stand-up motion; it must create clearance while building stable support.", "In simulation beneath lift-only plates, recovery reached 88.5 percent versus 18.9 percent without the geometric guidance; beneath freely moving randomized plates it reached 98.6 percent." ], "importance": 8, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-013/the-fallen-humanoid-learns-to-make-room-first", "json_url": "https://themachinepress.com/story/mp-2026-11-02-013.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-013-001", "text": "ROOF rewards a robot for reducing overlap with an overhead obstacle before standing, lifting simulated recovery from 18.9 to 88.5 percent in one setting.", "source_ids": [ "source-2026-11-02-013" ], "qualification": "A humanoid trapped beneath furniture cannot simply execute a standard stand-up motion; it must create clearance while building stable support." } ], "source_ids": [ "source-2026-11-02-013" ], "tags": [ "humanoid robots", "fall recovery", "reinforcement learning", "contact" ], "image_url": "https://themachinepress.com/issues/2026-11-02/robot-recovery.webp", "corrections": [] }, { "story_id": "mp-2026-11-02-014", "source_story_id": "tmp-story-drying-honeycomb-folds", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 14, "story_type": "dispatch", "section": "research", "editorial_classification": "editorial", "headline": "Drying Films Fold Into Open Honeycombs", "slug": "drying-films-fold-into-open-honeycombs", "dek": "Experiments and theory link the wavelength and arm length of three-way fold networks to dried-film thickness with the same five-eighths power law.", "summary": "Experiments and theory link the wavelength and arm length of three-way fold networks to dried-film thickness with the same five-eighths power law.", "body_text": "Solvent extraction and drying can turn a pinned viscoelastic film or gel into a repeating network of three-armed folds. The study attributes the pattern to compression in a stiffening, diffusively thickening surface skin above a drained underlayer. Both fold wavelength and arm length scaled with dried-film thickness to the five-eighths power and depended only weakly on elastic modulus. The folds met at 120-degree junctions on a triangular lattice, forming an open honeycomb network. Because they absorbed less than 0.1 percent of stored compressive strain, the authors point to elastocapillary arrest rather than bulk strain release as the limit on their amplitude. The mechanism is specific to the tested soft-film systems.", "why_it_matters": "Experiments and theory link the wavelength and arm length of three-way fold networks to dried-film thickness with the same five-eighths power law.", "limitations": [ "Both fold wavelength and arm length scaled with dried-film thickness to the five-eighths power and depended only weakly on elastic modulus." ], "importance": 7, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-014/drying-films-fold-into-open-honeycombs", "json_url": "https://themachinepress.com/story/mp-2026-11-02-014.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-014-001", "text": "Experiments and theory link the wavelength and arm length of three-way fold networks to dried-film thickness with the same five-eighths power law.", "source_ids": [ "source-2026-11-02-014" ], "qualification": "Both fold wavelength and arm length scaled with dried-film thickness to the five-eighths power and depended only weakly on elastic modulus." } ], "source_ids": [ "source-2026-11-02-014" ], "tags": [ "soft matter", "thin films", "pattern formation", "drying" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-026", "source_story_id": "tmp-story-thorium-clock-ultraviolet-source", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 15, "story_type": "dispatch", "section": "research", "editorial_classification": "editorial", "headline": "A Crystal Source Reaches the Nuclear Clock's Wavelength", "slug": "a-crystal-source-reaches-the-nuclear-clock-s-wavelength", "dek": "Random quasi-phase matching in strontium tetraborate produced continuous-wave 148.4-nanometer light, with output rising from 1.9 to 5.7 nanowatts under temperature optimization.", "summary": "Random quasi-phase matching in strontium tetraborate produced continuous-wave 148.4-nanometer light, with output rising from 1.9 to 5.7 nanowatts under temperature optimization.", "body_text": "Driving the thorium-229 isomer transition in a solid-state nuclear clock requires continuous-wave light near 148.4 nanometers. The reported source frequency-doubles 296.8-nanometer light in strontium tetraborate using random quasi-phase matching, then separates and measures the harmonic with a purpose-built prism spectrometer. Output reached 1.9 nanowatts at room temperature and 5.7 nanowatts at the optimum crystal temperature. The conversion response showed several closely spaced maxima between 30 and 50 degrees Celsius, so a small temperature change could move the source from a peak to a minimum. The work demonstrates and characterizes the source; it does not report a completed clock.", "why_it_matters": "Random quasi-phase matching in strontium tetraborate produced continuous-wave 148.4-nanometer light, with output rising from 1.9 to 5.7 nanowatts under temperature optimization.", "limitations": [ "The work demonstrates and characterizes the source; it does not report a completed clock." ], "importance": 8, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-026/a-crystal-source-reaches-the-nuclear-clock-s-wavelength", "json_url": "https://themachinepress.com/story/mp-2026-11-02-026.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-026-001", "text": "Random quasi-phase matching in strontium tetraborate produced continuous-wave 148.4-nanometer light, with output rising from 1.9 to 5.7 nanowatts under temperature optimization.", "source_ids": [ "source-2026-11-02-015" ], "qualification": "The work demonstrates and characterizes the source; it does not report a completed clock." } ], "source_ids": [ "source-2026-11-02-015" ], "tags": [ "nuclear clocks", "ultraviolet light", "nonlinear optics", "thorium-229" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-027", "source_story_id": "tmp-story-exoplanet-oxygen-o4-bands", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 16, "story_type": "dispatch", "section": "research", "editorial_classification": "editorial", "headline": "The Oxygen Question Needs the O4 Bands", "slug": "the-oxygen-question-needs-the-o4-bands", "dek": "Retrievals could mistake an Earth analog for an oxygen-dominated atmosphere unless near-infrared oxygen-pair absorption remained in the model and observing band.", "summary": "Retrievals could mistake an Earth analog for an oxygen-dominated atmosphere unless near-infrared oxygen-pair absorption remained in the model and observing band.", "body_text": "Reflected-light spectra can reveal oxygen on an Earth-sized exoplanet, but oxygen alone does not identify biology and retrievals can confuse composition with clouds, surface brightness and gravity. Simulations with the ExoReL retrieval framework found that omitting oxygen-oxygen collision-induced absorption allowed a modern Earth analog to be fit as an incorrect oxygen-dominated atmosphere. Including the O4 bands recovered the nitrogen-dominated, oxygen-rich solution without a tight mass prior. Under the authors' robustness rule, both tested observing configurations needed signal-to-noise of roughly 20 at 0.75 micrometers; ending near-infrared coverage at 1.4 instead of about 1.7 micrometers prevented recovery across the explored range. These are simulated retrieval requirements, not an observation of an exoplanet atmosphere.", "why_it_matters": "Retrievals could mistake an Earth analog for an oxygen-dominated atmosphere unless near-infrared oxygen-pair absorption remained in the model and observing band.", "limitations": [ "Reflected-light spectra can reveal oxygen on an Earth-sized exoplanet, but oxygen alone does not identify biology and retrievals can confuse composition with clouds, surface brightness and gravity.", "These are simulated retrieval requirements, not an observation of an exoplanet atmosphere." ], "importance": 8, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-027/the-oxygen-question-needs-the-o4-bands", "json_url": "https://themachinepress.com/story/mp-2026-11-02-027.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-027-001", "text": "Retrievals could mistake an Earth analog for an oxygen-dominated atmosphere unless near-infrared oxygen-pair absorption remained in the model and observing band.", "source_ids": [ "source-2026-11-02-016" ], "qualification": "Reflected-light spectra can reveal oxygen on an Earth-sized exoplanet, but oxygen alone does not identify biology and retrievals can confuse composition with clouds, surface brightness and gravity." } ], "source_ids": [ "source-2026-11-02-016" ], "tags": [ "exoplanets", "atmospheres", "spectroscopy", "biosignatures" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-015", "source_story_id": "tmp-brief-deferred-gate-recipient", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 17, "story_type": "ticker", "section": "research", "editorial_classification": "editorial", "headline": "A Quantum Gate Chooses Its Recipient Later", "slug": "a-quantum-gate-chooses-its-recipient-later", "dek": "A two-qudit entangling gate can be split among candidate nodes before one later key release selects the only node allowed to use it.", "summary": "A two-qudit entangling gate can be split among candidate nodes before one later key release selects the only node allowed to use it.", "body_text": "The protocol encrypts half of a Bell pair into shares and preserves secrecy against fewer than all candidate nodes. The selected node can apply the exact non-local gate; the key cannot be reused.", "why_it_matters": "A two-qudit entangling gate can be split among candidate nodes before one later key release selects the only node allowed to use it.", "limitations": [ "The selected node can apply the exact non-local gate; the key cannot be reused." ], "importance": 7, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-015/a-quantum-gate-chooses-its-recipient-later", "json_url": "https://themachinepress.com/story/mp-2026-11-02-015.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-015-001", "text": "A two-qudit entangling gate can be split among candidate nodes before one later key release selects the only node allowed to use it.", "source_ids": [ "source-2026-11-02-017" ], "qualification": "The selected node can apply the exact non-local gate; the key cannot be reused." } ], "source_ids": [ "source-2026-11-02-017" ], "tags": [ "quantum networks", "secret sharing", "gate teleportation" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-016", "source_story_id": "tmp-brief-instantaneous-kerr-response", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 18, "story_type": "ticker", "section": "research", "editorial_classification": "editorial", "headline": "The Kerr Response Is Measured in the Field", "slug": "the-kerr-response-is-measured-in-the-field", "dek": "Field-resolved pump-probe spectroscopy separated germanium's instantaneous nonlinear phase shift from delayed carrier effects at five micrometers.", "summary": "Field-resolved pump-probe spectroscopy separated germanium's instantaneous nonlinear phase shift from delayed carrier effects at five micrometers.", "body_text": "The measured phase followed the pump envelope and vanished when the pulses no longer overlapped. Its linear intensity scaling yielded a nonlinear refractive index of 0.87 plus or minus 0.16 times 10^-17 square meters per watt.", "why_it_matters": "Field-resolved pump-probe spectroscopy separated germanium's instantaneous nonlinear phase shift from delayed carrier effects at five micrometers.", "limitations": [], "importance": 7, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-016/the-kerr-response-is-measured-in-the-field", "json_url": "https://themachinepress.com/story/mp-2026-11-02-016.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-016-001", "text": "Field-resolved pump-probe spectroscopy separated germanium's instantaneous nonlinear phase shift from delayed carrier effects at five micrometers.", "source_ids": [ "source-2026-11-02-018" ], "qualification": null } ], "source_ids": [ "source-2026-11-02-018" ], "tags": [ "photonics", "Kerr effect", "semiconductors" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-017", "source_story_id": "tmp-brief-ice-lake-convection-onset", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 19, "story_type": "ticker", "section": "research", "editorial_classification": "editorial", "headline": "Ice-Covered Lakes Wait for Instability", "slug": "ice-covered-lakes-wait-for-instability", "dek": "A linear model and direct simulations connect thermal stratification, water clarity and sunlight to the onset of radiatively driven convection.", "summary": "A linear model and direct simulations connect thermal stratification, water clarity and sunlight to the onset of radiatively driven convection.", "body_text": "Weakly stratified lakes could become convective within hours, while strongly stratified cases could remain dormant after a full clear-sky spring day. The linear analysis matched the simulated early instability growth.", "why_it_matters": "A linear model and direct simulations connect thermal stratification, water clarity and sunlight to the onset of radiatively driven convection.", "limitations": [ "Weakly stratified lakes could become convective within hours, while strongly stratified cases could remain dormant after a full clear-sky spring day." ], "importance": 6, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-017/ice-covered-lakes-wait-for-instability", "json_url": "https://themachinepress.com/story/mp-2026-11-02-017.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-017-001", "text": "A linear model and direct simulations connect thermal stratification, water clarity and sunlight to the onset of radiatively driven convection.", "source_ids": [ "source-2026-11-02-019" ], "qualification": "Weakly stratified lakes could become convective within hours, while strongly stratified cases could remain dormant after a full clear-sky spring day." } ], "source_ids": [ "source-2026-11-02-019" ], "tags": [ "lakes", "convection", "climate physics" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-018", "source_story_id": "tmp-brief-soft-xray-ilgad", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 20, "story_type": "ticker", "section": "chips-infrastructure", "editorial_classification": "editorial", "headline": "A Pixel Sensor Resolves 390-Electron-Volt Photons", "slug": "a-pixel-sensor-resolves-390-electron-volt-photons", "dek": "A thin-window iLGAD sensor achieved single-photon resolution below the nitrogen K-edge with a signal-to-noise ratio of five.", "summary": "A thin-window iLGAD sensor achieved single-photon resolution below the nitrogen K-edge with a signal-to-noise ratio of five.", "body_text": "Two gain-layer designs exposed a double-peak spectrum tied to absorption depth. A fitted model then estimated the entrance and gain-layer geometry, informing larger high-frame-rate soft-X-ray detectors.", "why_it_matters": "A thin-window iLGAD sensor achieved single-photon resolution below the nitrogen K-edge with a signal-to-noise ratio of five.", "limitations": [], "importance": 7, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-018/a-pixel-sensor-resolves-390-electron-volt-photons", "json_url": "https://themachinepress.com/story/mp-2026-11-02-018.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-018-001", "text": "A thin-window iLGAD sensor achieved single-photon resolution below the nitrogen K-edge with a signal-to-noise ratio of five.", "source_ids": [ "source-2026-11-02-020" ], "qualification": null } ], "source_ids": [ "source-2026-11-02-020" ], "tags": [ "X-ray detectors", "sensors", "synchrotrons" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-019", "source_story_id": "tmp-brief-tess-blend-ceiling", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 21, "story_type": "ticker", "section": "research", "editorial_classification": "editorial", "headline": "A Transit Classifier Meets the Blend Ceiling", "slug": "a-transit-classifier-meets-the-blend-ceiling", "dek": "A TESS classifier's ROC area fell from 0.97 to 0.50 against real rejected candidates; a multi-stage pipeline then independently validated TOI-6662 b.", "summary": "A TESS classifier's ROC area fell from 0.97 to 0.50 against real rejected candidates; a multi-stage pipeline then independently validated TOI-6662 b.", "body_text": "Blended eclipses made 42 percent of 145 examined false positives and capped what one light curve could reveal. The pipeline added centroid, scene and Gaia evidence before validating a 0.94-Earth-radius planet on a 7.05-hour orbit.", "why_it_matters": "A TESS classifier's ROC area fell from 0.97 to 0.50 against real rejected candidates; a multi-stage pipeline then independently validated TOI-6662 b.", "limitations": [], "importance": 7, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-019/a-transit-classifier-meets-the-blend-ceiling", "json_url": "https://themachinepress.com/story/mp-2026-11-02-019.json", "first_published_at": "2026-11-02T09:00:00.000-05:00", "modified_at": "2026-11-02T09:00:00.000-05:00", "content_status": "new", "is_carryover": false, "carryover_reason": null, "key_claims": [ { "claim_id": "claim-mp-2026-11-02-019-001", "text": "A TESS classifier's ROC area fell from 0.97 to 0.50 against real rejected candidates; a multi-stage pipeline then independently validated TOI-6662 b.", "source_ids": [ "source-2026-11-02-021" ], "qualification": null } ], "source_ids": [ "source-2026-11-02-021" ], "tags": [ "exoplanets", "TESS", "computer vision", "validation" ], "image_url": null, "corrections": [] }, { "story_id": "mp-2026-11-02-020", "source_story_id": "invention-space-drums-2", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 22, "story_type": "invention_desk", "section": "invention-desk", "editorial_classification": "editorial", "headline": "Space Drums 2.0", "slug": "space-drums-2-0", "dek": "Moves motion tracking onto two ESP32-S3 drumsticks, adds haptic hits and a two-pedal hub, and sends velocity-sensitive drum events to open PC or Android playback software.", "summary": "Moves motion tracking onto two ESP32-S3 drumsticks, adds haptic hits and a two-pedal hub, and sends velocity-sensitive drum events to open PC or Android playback software.", "body_text": "Moves motion tracking onto two ESP32-S3 drumsticks, adds haptic hits and a two-pedal hub, and sends velocity-sensitive drum events to open PC or Android playback software.", "why_it_matters": "An independent builder is turning an improbable idea into a working project.", "limitations": [ "A Desk Pick is an editorial selection, not a product endorsement." ], "importance": null, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-020/space-drums-2-0", "json_url": "https://themachinepress.com/story/mp-2026-11-02-020.json", "first_published_at": "2026-09-27T09:00:00.000-04:00", "modified_at": "2026-09-27T09:00:00.000-04:00", "content_status": "carried_over", "is_carryover": true, "carryover_reason": "invention_desk_seven_day_placement", "key_claims": [], "source_ids": [], "tags": [ "independent-builders", "desk-pick", "released" ], "image_url": "https://themachinepress.com/issues/2026-11-02/invention-desk/space-drums-2.png", "corrections": [] }, { "story_id": "mp-2026-11-02-021", "source_story_id": "invention-3ric", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 23, "story_type": "invention_desk", "section": "invention-desk", "editorial_classification": "editorial", "headline": "3RIC", "slug": "3ric", "dek": "Pairs a chip-level 65C02 computer with custom hardware, video, sound, floppy, and microSD support with a cycle-honest emulator and browser assembler built around the same machine.", "summary": "Pairs a chip-level 65C02 computer with custom hardware, video, sound, floppy, and microSD support with a cycle-honest emulator and browser assembler built around the same machine.", "body_text": "Pairs a chip-level 65C02 computer with custom hardware, video, sound, floppy, and microSD support with a cycle-honest emulator and browser assembler built around the same machine.", "why_it_matters": "An independent builder is turning an improbable idea into a working project.", "limitations": [ "A Desk Pick is an editorial selection, not a product endorsement." ], "importance": null, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-021/3ric", "json_url": "https://themachinepress.com/story/mp-2026-11-02-021.json", "first_published_at": "2026-09-27T09:00:00.000-04:00", "modified_at": "2026-09-27T09:00:00.000-04:00", "content_status": "carried_over", "is_carryover": true, "carryover_reason": "invention_desk_seven_day_placement", "key_claims": [], "source_ids": [], "tags": [ "independent-builders", "desk-pick", "released" ], "image_url": "https://themachinepress.com/issues/2026-11-02/invention-desk/3ric.png", "corrections": [] }, { "story_id": "mp-2026-11-02-022", "source_story_id": "invention-m3-crete", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 24, "story_type": "invention_desk", "section": "invention-desk", "editorial_classification": "editorial", "headline": "M3-CRETE", "slug": "m3-crete", "dek": "Publishes a pallet-scale cartesian motion reference, CAD, BOM, and low-voltage controls for supervised cementitious-extrusion research while explicitly excluding certified structural use.", "summary": "Publishes a pallet-scale cartesian motion reference, CAD, BOM, and low-voltage controls for supervised cementitious-extrusion research while explicitly excluding certified structural use.", "body_text": "Publishes a pallet-scale cartesian motion reference, CAD, BOM, and low-voltage controls for supervised cementitious-extrusion research while explicitly excluding certified structural use.", "why_it_matters": "An independent builder is turning an improbable idea into a working project.", "limitations": [ "A Desk Pick is an editorial selection, not a product endorsement." ], "importance": null, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-022/m3-crete", "json_url": "https://themachinepress.com/story/mp-2026-11-02-022.json", "first_published_at": "2026-09-27T09:00:00.000-04:00", "modified_at": "2026-09-27T09:00:00.000-04:00", "content_status": "carried_over", "is_carryover": true, "carryover_reason": "invention_desk_seven_day_placement", "key_claims": [], "source_ids": [], "tags": [ "independent-builders", "desk-pick", "prototype" ], "image_url": "https://themachinepress.com/issues/2026-11-02/invention-desk/m3-crete.png", "corrections": [] }, { "story_id": "mp-2026-11-02-023", "source_story_id": "invention-desknode", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 25, "story_type": "invention_desk", "section": "invention-desk", "editorial_classification": "editorial", "headline": "DeskNode", "slug": "desknode", "dek": "Turns a bare ESP32-S3 touchscreen into a USB-connected desk instrument for PC activity and optional room temperature, humidity, and light measurements.", "summary": "Turns a bare ESP32-S3 touchscreen into a USB-connected desk instrument for PC activity and optional room temperature, humidity, and light measurements.", "body_text": "Turns a bare ESP32-S3 touchscreen into a USB-connected desk instrument for PC activity and optional room temperature, humidity, and light measurements.", "why_it_matters": "An independent builder is turning an improbable idea into a working project.", "limitations": [ "A Desk Pick is an editorial selection, not a product endorsement." ], "importance": null, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-023/desknode", "json_url": "https://themachinepress.com/story/mp-2026-11-02-023.json", "first_published_at": "2026-09-27T09:00:00.000-04:00", "modified_at": "2026-09-27T09:00:00.000-04:00", "content_status": "carried_over", "is_carryover": true, "carryover_reason": "invention_desk_seven_day_placement", "key_claims": [], "source_ids": [], "tags": [ "independent-builders", "desk-pick", "prototype" ], "image_url": "https://themachinepress.com/issues/2026-11-02/invention-desk/desknode.png", "corrections": [] }, { "story_id": "mp-2026-11-02-024", "source_story_id": "invention-sponsored-house-example", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 26, "story_type": "invention_desk", "section": "invention-desk", "editorial_classification": "house_example", "headline": "The First Paid Slot", "slug": "the-first-paid-slot", "dek": "A transparent preview of paid placement with one verified link and no claim of endorsement.", "summary": "A transparent preview of paid placement with one verified link and no claim of endorsement.", "body_text": "A transparent preview of paid placement with one verified link and no claim of endorsement.\n\nHouse example - no advertiser paid. Payment will buy placement, never endorsement.", "why_it_matters": "This placement explains how builders can appear in The Invention Desk without purchasing editorial endorsement.", "limitations": [ "House example - no advertiser paid. Payment will buy placement, never endorsement." ], "importance": null, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-024/the-first-paid-slot", "json_url": "https://themachinepress.com/story/mp-2026-11-02-024.json", "first_published_at": "2026-07-10T09:00:00.000-04:00", "modified_at": "2026-07-10T09:00:00.000-04:00", "content_status": "carried_over", "is_carryover": true, "carryover_reason": "invention_desk_seven_day_placement", "key_claims": [], "source_ids": [], "tags": [ "independent-builders", "sponsored-project", "open" ], "image_url": "https://themachinepress.com/issues/2026-11-02/invention-desk/sponsored-project.webp", "corrections": [] }, { "story_id": "mp-2026-11-02-025", "source_story_id": "invention-placement-cta", "edition_id": "mp-2026-11-02-morning-0115", "edition_url": "https://themachinepress.com/edition/2026-11-02", "position": 27, "story_type": "invention_desk", "section": "invention-desk", "editorial_classification": "house_example", "headline": "Put Your Project on the Desk", "slug": "put-your-project-on-the-desk", "dek": "One manually reviewed placement stays active for seven days and remains separate from Desk Picks.", "summary": "One manually reviewed placement stays active for seven days and remains separate from Desk Picks.", "body_text": "One manually reviewed placement stays active for seven days and remains separate from Desk Picks.\n\nManual intake only. Payment buys placement, never endorsement, and every submission is reviewed.", "why_it_matters": "This placement explains how builders can appear in The Invention Desk without purchasing editorial endorsement.", "limitations": [ "Manual intake only. Payment buys placement, never endorsement, and every submission is reviewed." ], "importance": null, "canonical_url": "https://themachinepress.com/story/mp-2026-11-02-025/put-your-project-on-the-desk", "json_url": "https://themachinepress.com/story/mp-2026-11-02-025.json", "first_published_at": "2026-07-10T09:00:00.000-04:00", "modified_at": "2026-07-10T09:00:00.000-04:00", "content_status": "carried_over", "is_carryover": true, "carryover_reason": "invention_desk_seven_day_placement", "key_claims": [], "source_ids": [], "tags": [ "independent-builders", "cta", "open" ], "image_url": "https://themachinepress.com/issues/2026-11-02/invention-desk/put-your-project-on-the-desk.webp", "corrections": [] } ], "sources": [ { "source_id": "source-2026-11-02-001", "title": "arXiv preprint 2610.42152, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42152v1", "canonical_url": "https://arxiv.org/abs/2610.42152v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:59:59.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-001-001" ] }, { "source_id": "source-2026-11-02-002", "title": "arXiv preprint 2610.42141, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42141v1", "canonical_url": "https://arxiv.org/abs/2610.42141v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:58:48.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-002-001" ] }, { "source_id": "source-2026-11-02-003", "title": "arXiv preprint 2610.42149, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42149v1", "canonical_url": "https://arxiv.org/abs/2610.42149v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:59:43.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-003-001" ] }, { "source_id": "source-2026-11-02-004", "title": "arXiv preprint 2610.42148, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42148v1", "canonical_url": "https://arxiv.org/abs/2610.42148v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:59:38.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-004-001" ] }, { "source_id": "source-2026-11-02-005", "title": "arXiv preprint 2610.42147, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42147v1", "canonical_url": "https://arxiv.org/abs/2610.42147v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:59:34.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-005-001" ] }, { "source_id": "source-2026-11-02-006", "title": "arXiv preprint 2610.42146, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42146v1", "canonical_url": "https://arxiv.org/abs/2610.42146v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:59:25.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-006-001" ] }, { "source_id": "source-2026-11-02-007", "title": "arXiv preprint 2610.42144, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42144v1", "canonical_url": "https://arxiv.org/abs/2610.42144v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:59:12.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-007-001" ] }, { "source_id": "source-2026-11-02-008", "title": "arXiv preprint 2610.42139, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42139v1", "canonical_url": "https://arxiv.org/abs/2610.42139v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:58:42.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-008-001" ] }, { "source_id": "source-2026-11-02-009", "title": "arXiv preprint 2610.42136, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42136v1", "canonical_url": "https://arxiv.org/abs/2610.42136v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:58:33.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-009-001" ] }, { "source_id": "source-2026-11-02-010", "title": "arXiv preprint 2610.42130, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42130v1", "canonical_url": "https://arxiv.org/abs/2610.42130v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:57:25.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-010-001" ] }, { "source_id": "source-2026-11-02-011", "title": "arXiv preprint 2610.42121, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42121v1", "canonical_url": "https://arxiv.org/abs/2610.42121v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:53:28.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-011-001" ] }, { "source_id": "source-2026-11-02-012", "title": "arXiv preprint 2610.42115, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42115v1", "canonical_url": "https://arxiv.org/abs/2610.42115v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:49:52.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-012-001" ] }, { "source_id": "source-2026-11-02-013", "title": "arXiv preprint 2610.42097, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42097v1", "canonical_url": "https://arxiv.org/abs/2610.42097v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:41:32.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-013-001" ] }, { "source_id": "source-2026-11-02-014", "title": "arXiv preprint 2610.42074, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42074v1", "canonical_url": "https://arxiv.org/abs/2610.42074v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:31:19.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-014-001" ] }, { "source_id": "source-2026-11-02-015", "title": "arXiv preprint 2610.42055, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42055v1", "canonical_url": "https://arxiv.org/abs/2610.42055v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:18:40.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-026-001" ] }, { "source_id": "source-2026-11-02-016", "title": "arXiv preprint 2610.42035, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42035v1", "canonical_url": "https://arxiv.org/abs/2610.42035v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:01:07.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-027-001" ] }, { "source_id": "source-2026-11-02-017", "title": "arXiv preprint 2610.42150, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42150v1", "canonical_url": "https://arxiv.org/abs/2610.42150v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:59:43.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-015-001" ] }, { "source_id": "source-2026-11-02-018", "title": "arXiv preprint 2610.42098, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42098v1", "canonical_url": "https://arxiv.org/abs/2610.42098v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:42:49.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-016-001" ] }, { "source_id": "source-2026-11-02-019", "title": "arXiv preprint 2610.42089, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42089v1", "canonical_url": "https://arxiv.org/abs/2610.42089v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:38:01.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-017-001" ] }, { "source_id": "source-2026-11-02-020", "title": "arXiv preprint 2610.42064, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42064v1", "canonical_url": "https://arxiv.org/abs/2610.42064v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T13:25:31.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-018-001" ] }, { "source_id": "source-2026-11-02-021", "title": "arXiv preprint 2610.42009, original v1", "publisher": "arXiv", "url": "https://arxiv.org/abs/2610.42009v1", "canonical_url": "https://arxiv.org/abs/2610.42009v1", "source_type": "primary_research", "is_primary_source": true, "published_at": "2026-10-30T12:47:34.000-04:00", "accessed_at": "2026-11-02T08:25:20.000-05:00", "supports_claim_ids": [ "claim-mp-2026-11-02-019-001" ] } ], "corrections": [], "publisher": { "name": "The Machine Press", "url": "https://themachinepress.com", "description": "A daily newspaper for the age of artificial intelligence." } }