--- schema_version: "1.0.0" edition_id: "mp-2026-09-25-morning-0078" published_at: "2026-09-25T09:00:00.000-04:00" modified_at: "2026-09-25T09:00:00.000-04:00" canonical_url: "https://themachinepress.com/edition/2026-09-25" story_count: 27 lead_story_id: "mp-2026-09-25-001" --- # The Machine Press — Morning edition Edition ID: `mp-2026-09-25-morning-0078` Published: 2026-09-25T09:00:00.000-04:00 Canonical edition: https://themachinepress.com/edition/2026-09-25 A cryogenic antenna collected 97% of the emitter's light while preserving high purity and indistinguishability. ## 1. One Molecule Delivered 100 Million Photons a Second {#mp-2026-09-25-001} - Story ID: `mp-2026-09-25-001` - Type: `lead` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-001/one-molecule-delivered-100-million-photons-a-second **Dek:** A cryogenic antenna collected 97% of the emitter's light while preserving high purity and indistinguishability. Researchers integrated a single dibenzoterrylene molecule with a planar metallo-dielectric antenna and directed its cryogenic emission into collection optics. The device delivered 100 million photons per second into the first lens with 97% collection efficiency, while the authors measured 91.2% photon indistinguishability and 97.7% single-photon purity under continuous-wave and pulsed excitation. The result is a laboratory demonstration from one organic emitter, not yet a scalable quantum-photonic system, but it combines rate, collection and photon quality in one molecular source. ### Why it matters {#why-it-matters-mp-2026-09-25-001} A cryogenic antenna collected 97% of the emitter's light while preserving high purity and indistinguishability. ### Limits and context {#limitations-mp-2026-09-25-001} - The result is a laboratory demonstration from one organic emitter, not yet a scalable quantum-photonic system, but it combines rate, collection and photon quality in one molecular source. ### Claims and sources {#claims-mp-2026-09-25-001} - A cryogenic antenna collected 97% of the emitter's light while preserving high purity and indistinguishability. [source-2026-09-25-001] — Qualification: The result is a laboratory demonstration from one organic emitter, not yet a scalable quantum-photonic system, but it combines rate, collection and photon quality in one molecular source. ## 2. Simulated Customers Screened an AI Agent Before Launch {#mp-2026-09-25-002} - Story ID: `mp-2026-09-25-002` - Type: `secondary` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-002/simulated-customers-screened-an-ai-agent-before-launch **Dek:** Nubank reports a 36.69-point tNPS gain in one live test and an 8.82-point self-service gain in another. A team used synthetic customers and simulated tool responses to rehearse customer-service agents without touching production backends. Across four deployed versions of Nubank's card-support agents, version-level simulator scores tracked production evaluator scores. Simulation-guided iteration preceded a live A/B test that raised transactional net promoter score by 36.69 points; a separate screen of more than 16,000 simulated conversations selected an open-weight configuration that increased self-service by 8.82 percentage points without a statistically significant tNPS change. These are company-specific results reported by the authors, not proof that simulated testing predicts every production environment. ### Why it matters {#why-it-matters-mp-2026-09-25-002} Nubank reports a 36.69-point tNPS gain in one live test and an 8.82-point self-service gain in another. ### Limits and context {#limitations-mp-2026-09-25-002} - These are company-specific results reported by the authors, not proof that simulated testing predicts every production environment. ### Claims and sources {#claims-mp-2026-09-25-002} - Nubank reports a 36.69-point tNPS gain in one live test and an 8.82-point self-service gain in another. [source-2026-09-25-002] — Qualification: These are company-specific results reported by the authors, not proof that simulated testing predicts every production environment. ## 3. Conversation Drift Did Not Make Secrets Fade {#mp-2026-09-25-003} - Story ID: `mp-2026-09-25-003` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-003/conversation-drift-did-not-make-secrets-fade **Dek:** PrivDrift recovered seeded secrets in 38.7% to 54.6% of tested long-context dialogues. PrivDrift placed seeded secrets into 1,000 controlled multi-turn conversations, shifted the topic, then applied standardized persuasion and extraction probes. Across three extended-context models, the authors report dialogue-level hybrid leakage between 38.7% and 54.6%. More topic drift within the tested window did not reliably reduce recovery, framing active-session privacy as a persistent behavioral risk rather than only a training-data or immediate-jailbreak problem. ### Why it matters {#why-it-matters-mp-2026-09-25-003} PrivDrift recovered seeded secrets in 38.7% to 54.6% of tested long-context dialogues. ### Limits and context {#limitations-mp-2026-09-25-003} - More topic drift within the tested window did not reliably reduce recovery, framing active-session privacy as a persistent behavioral risk rather than only a training-data or immediate-jailbreak problem. ### Claims and sources {#claims-mp-2026-09-25-003} - PrivDrift recovered seeded secrets in 38.7% to 54.6% of tested long-context dialogues. [source-2026-09-25-003] — Qualification: More topic drift within the tested window did not reliably reduce recovery, framing active-session privacy as a persistent behavioral risk rather than only a training-data or immediate-jailbreak problem. ## 4. DKIST Resolved Solar Vortices Smaller Than 50 Kilometers {#mp-2026-09-25-004} - Story ID: `mp-2026-09-25-004` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-004/dkist-resolved-solar-vortices-smaller-than-50-kilometers **Dek:** The 4-meter telescope yielded 201 detections with a median equivalent diameter of 38.6 kilometers. High-resolution 416-nanometer observations from the Daniel K. Inouye Solar Telescope covered 143 reconstructed frames over about 3.2 minutes. The analysis identified 201 photospheric vortex detections with a median equivalent diameter of 38.6 kilometers and median swirling period of 37.3 seconds. Virtually all were below 50 kilometers, matching a scale predicted by magnetohydrodynamic simulations but previously beyond direct observational reach, according to the authors. ### Why it matters {#why-it-matters-mp-2026-09-25-004} The 4-meter telescope yielded 201 detections with a median equivalent diameter of 38.6 kilometers. ### Limits and context {#limitations-mp-2026-09-25-004} - Virtually all were below 50 kilometers, matching a scale predicted by magnetohydrodynamic simulations but previously beyond direct observational reach, according to the authors. ### Claims and sources {#claims-mp-2026-09-25-004} - The 4-meter telescope yielded 201 detections with a median equivalent diameter of 38.6 kilometers. [source-2026-09-25-004] — Qualification: Virtually all were below 50 kilometers, matching a scale predicted by magnetohydrodynamic simulations but previously beyond direct observational reach, according to the authors. ## 5. Physicians Mistook Synthetic Charts for Real Ones {#mp-2026-09-25-005} - Story ID: `mp-2026-09-25-005` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-005/physicians-mistook-synthetic-charts-for-real-ones **Dek:** The open benchmark contains 1,268 longitudinal patients and 5,602 fully synthetic encounters. Synthetic Hospital builds longitudinal electronic-health-record cases from public medical-education material, with ontology-grounded diagnoses, findings and time relations rather than protected patient data. In a blinded review, physicians distinguished its charts from real records at a near-chance 53%. The best of ten tested models scored 0.73 severity-weighted F1 on reconstructing problem lists—matching the seven-physician mean on a subset but below the best physician's 0.89—and missed roughly half of clinically relevant findings in chart summaries. ### Why it matters {#why-it-matters-mp-2026-09-25-005} The open benchmark contains 1,268 longitudinal patients and 5,602 fully synthetic encounters. ### Limits and context {#limitations-mp-2026-09-25-005} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-09-25-005} - The open benchmark contains 1,268 longitudinal patients and 5,602 fully synthetic encounters. [source-2026-09-25-005] ## 6. A Dark-Galaxy Candidate Appeared Without Starlight {#mp-2026-09-25-006} - Story ID: `mp-2026-09-25-006` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-006/a-dark-galaxy-candidate-appeared-without-starlight **Dek:** FAST found compact hydrogen, but deep optical imaging found no counterpart down to a faint surface-brightness limit. A blind FAST hydrogen survey and DESI Legacy Survey imaging identified J1351+0039 as an isolated Reionization-Limited H I Cloud candidate. The object has an inferred neutral-hydrogen mass of 480,000 solar masses, no detected optical counterpart to a g-band surface-brightness limit of 28.1 magnitudes per square arcsecond, and an estimated dynamical mass of 130 million solar masses. The authors derive a dark-to-baryonic mass ratio near 200, while describing the classification as a promising candidate rather than a settled dark-galaxy discovery. ### Why it matters {#why-it-matters-mp-2026-09-25-006} FAST found compact hydrogen, but deep optical imaging found no counterpart down to a faint surface-brightness limit. ### Limits and context {#limitations-mp-2026-09-25-006} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-09-25-006} - FAST found compact hydrogen, but deep optical imaging found no counterpart down to a faint surface-brightness limit. [source-2026-09-25-006] ## 7. The Agent’s Completion Claim Beat Its Pass Rate by Up to 37.9 Points {#mp-2026-09-25-007} - Story ID: `mp-2026-09-25-007` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-007/the-agent-s-completion-claim-beat-its-pass-rate-by-up-to-37-9-points **Dek:** SpecHarness separates an agent's proposal from evidence that requirements were actually satisfied. Researchers extracted 509 source-grounded directions from SkillsBench and tested seven models. They report that 79.6% to 86.4% of directions were satisfied, while agents' completion-claim rates exceeded official evaluator pass rates by 28.7 to 37.9 percentage points. SpecHarness compiles visible specifications into source-linked obligations and lets only qualified evidence establish governed state, leaving ambiguous requirements advisory. The work argues for an authority boundary, not merely another instruction prompt. ### Why it matters {#why-it-matters-mp-2026-09-25-007} SpecHarness separates an agent's proposal from evidence that requirements were actually satisfied. ### Limits and context {#limitations-mp-2026-09-25-007} - SpecHarness compiles visible specifications into source-linked obligations and lets only qualified evidence establish governed state, leaving ambiguous requirements advisory. - The work argues for an authority boundary, not merely another instruction prompt. ### Claims and sources {#claims-mp-2026-09-25-007} - SpecHarness separates an agent's proposal from evidence that requirements were actually satisfied. [source-2026-09-25-007] — Qualification: SpecHarness compiles visible specifications into source-linked obligations and lets only qualified evidence establish governed state, leaving ambiguous requirements advisory. ## 8. A Robot Hand Recovered Its Grip Without Palm Tactile Sensors {#mp-2026-09-25-008} - Story ID: `mp-2026-09-25-008` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-008/a-robot-hand-recovered-its-grip-without-palm-tactile-sensors **Dek:** A controller recomputed forces as contact moved across a 27-degree-of-freedom arm and hand. The controller estimates whole-hand contacts from a tracked object model and proprioception, then repeatedly reallocates forces under friction, actuator and consistency constraints. In simulation it retained grasps better than fixed-allocation and fingertip-only baselines under controlled perturbations. Hardware experiments on a 27-degree-of-freedom arm-hand system showed grasp maintenance and regrasping after human-applied disturbances as contacts moved across the hand; the tests do not establish performance across arbitrary objects or untracked scenes. ### Why it matters {#why-it-matters-mp-2026-09-25-008} A controller recomputed forces as contact moved across a 27-degree-of-freedom arm and hand. ### Limits and context {#limitations-mp-2026-09-25-008} - In simulation it retained grasps better than fixed-allocation and fingertip-only baselines under controlled perturbations. - Hardware experiments on a 27-degree-of-freedom arm-hand system showed grasp maintenance and regrasping after human-applied disturbances as contacts moved across the hand; the tests do not establish performance across arbitrary objects or untracked scenes. ### Claims and sources {#claims-mp-2026-09-25-008} - A controller recomputed forces as contact moved across a 27-degree-of-freedom arm and hand. [source-2026-09-25-008] — Qualification: In simulation it retained grasps better than fixed-allocation and fingertip-only baselines under controlled perturbations. ## 9. A Learned Quantum Scheduler Held Success With 71% Weaker Links {#mp-2026-09-25-009} - Story ID: `mp-2026-09-25-009` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-009/a-learned-quantum-scheduler-held-success-with-71-weaker-links **Dek:** Message-passing reinforcement learning scheduled simultaneous entanglement requests across tested network topologies. The authors formulated simultaneous quantum-network requests as a Markov decision process and trained graph-based double deep-Q policies. On tested physically relevant topologies, the learned policies maintained 100% success at link-activation probabilities up to 71% lower than heuristic baselines. With hardware-placement restrictions, they retained at least 80% success at probabilities up to 59% lower. These are modeled network results, not deployed quantum-internet measurements. ### Why it matters {#why-it-matters-mp-2026-09-25-009} Message-passing reinforcement learning scheduled simultaneous entanglement requests across tested network topologies. ### Limits and context {#limitations-mp-2026-09-25-009} - These are modeled network results, not deployed quantum-internet measurements. ### Claims and sources {#claims-mp-2026-09-25-009} - Message-passing reinforcement learning scheduled simultaneous entanglement requests across tested network topologies. [source-2026-09-25-009] — Qualification: These are modeled network results, not deployed quantum-internet measurements. ## 10. The Excavator Lifted 6.52 Kilograms Per Cycle {#mp-2026-09-25-010} - Story ID: `mp-2026-09-25-010` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-010/the-excavator-lifted-6-52-kilograms-per-cycle **Dek:** A scaled hydraulic machine combined LiDAR target selection with learned approach, digging and lifting policies. A learning-based excavation stack selected targets from LiDAR elevation maps, used reinforcement learning for approach and loaded transport, and used imitation learning for vision-based digging and lifting. On a scaled hydraulic excavator, the learned digging policy averaged 6.52 kilograms per completed cycle versus 2.68 kilograms for a fixed-dig baseline. Three five-scoop runs demonstrated consecutive operation as the pile changed, but the evidence remains a scaled experimental system rather than full-size field deployment. ### Why it matters {#why-it-matters-mp-2026-09-25-010} A scaled hydraulic machine combined LiDAR target selection with learned approach, digging and lifting policies. ### Limits and context {#limitations-mp-2026-09-25-010} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-09-25-010} - A scaled hydraulic machine combined LiDAR target selection with learned approach, digging and lifting policies. [source-2026-09-25-010] ## 11. Pairwise Scoring Picked the Wrong Robot Plan on Six of Seven Maps {#mp-2026-09-25-011} - Story ID: `mp-2026-09-25-011` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-011/pairwise-scoring-picked-the-wrong-robot-plan-on-six-of-seven-maps **Dek:** Dropping three- and four-robot interactions produced regret as high as 0.337 of map coverage. Researchers replayed all 16 subsets of each four-robot exploration plan to calculate exact delivered coverage, then compared that ranking with scores built only from singleton and pair terms. At the 15-meter candidate range, the order-two approximation changed the selected plan on six of seven maps in each of two plan families, with regret up to 0.337 of total map coverage. A least-squares pairwise fit improved the result but still changed the winner on three of seven maps per family. ### Why it matters {#why-it-matters-mp-2026-09-25-011} Dropping three- and four-robot interactions produced regret as high as 0.337 of map coverage. ### Limits and context {#limitations-mp-2026-09-25-011} - Researchers replayed all 16 subsets of each four-robot exploration plan to calculate exact delivered coverage, then compared that ranking with scores built only from singleton and pair terms. ### Claims and sources {#claims-mp-2026-09-25-011} - Dropping three- and four-robot interactions produced regret as high as 0.337 of map coverage. [source-2026-09-25-011] — Qualification: Researchers replayed all 16 subsets of each four-robot exploration plan to calculate exact delivered coverage, then compared that ranking with scores built only from singleton and pair terms. ## 12. Syndrome Measurements Made Fault-Tolerant T Gates Deterministic {#mp-2026-09-25-012} - Story ID: `mp-2026-09-25-012` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-012/syndrome-measurements-made-fault-tolerant-t-gates-deterministic **Dek:** The construction applies to every stabilizer code of distance at least two in a suitable logical basis. The proposed mechanism releases one stabilizer check to expose an extra logical qubit, applies two Pauli rotations, and uses syndrome measurement plus Clifford feed-forward to implement a deterministic logical T gate. The authors give two single-fault-tolerant circuits: a fixed 22-qubit construction and a Golay-code gate. Serial implementations with ancilla reuse require at most 33 and 32 physical qubits, respectively. The result is a theoretical circuit construction, not a reported hardware demonstration. ### Why it matters {#why-it-matters-mp-2026-09-25-012} The construction applies to every stabilizer code of distance at least two in a suitable logical basis. ### Limits and context {#limitations-mp-2026-09-25-012} - The result is a theoretical circuit construction, not a reported hardware demonstration. ### Claims and sources {#claims-mp-2026-09-25-012} - The construction applies to every stabilizer code of distance at least two in a suitable logical basis. [source-2026-09-25-012] — Qualification: The result is a theoretical circuit construction, not a reported hardware demonstration. ## 13. A Humanoid Learned Motion Straight From Monocular Video {#mp-2026-09-25-013} - Story ID: `mp-2026-09-25-013` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-013/a-humanoid-learned-motion-straight-from-monocular-video **Dek:** BeyondRetarget maps RGB video directly to robot motion instead of first reconstructing a human skeleton. BeyondRetarget learns a robot-oriented representation directly from monocular RGB video, avoiding a separate human-motion reconstruction and retargeting stage. A contact-aware optimization step improves temporal consistency and physical plausibility. The authors report higher motion accuracy, execution success and robustness with lower latency in simulation and on real humanoid robots, but the abstract does not supply absolute success rates or establish performance outside the evaluated motions and platforms. ### Why it matters {#why-it-matters-mp-2026-09-25-013} BeyondRetarget maps RGB video directly to robot motion instead of first reconstructing a human skeleton. ### Limits and context {#limitations-mp-2026-09-25-013} - The authors report higher motion accuracy, execution success and robustness with lower latency in simulation and on real humanoid robots, but the abstract does not supply absolute success rates or establish performance outside the evaluated motions and platforms. ### Claims and sources {#claims-mp-2026-09-25-013} - BeyondRetarget maps RGB video directly to robot motion instead of first reconstructing a human skeleton. [source-2026-09-25-013] — Qualification: The authors report higher motion accuracy, execution success and robustness with lower latency in simulation and on real humanoid robots, but the abstract does not supply absolute success rates or establish performance outside the evaluated motions and platforms. ## 14. Stripped Binary Stars May Explain a Dwarf Galaxy’s Hard Light {#mp-2026-09-25-014} - Story ID: `mp-2026-09-25-014` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-014/stripped-binary-stars-may-explain-a-dwarf-galaxy-s-hard-light **Dek:** JWST and ALMA observations point to star formation—not an active galactic nucleus—in Pox 186. JWST mid-infrared spectroscopy and archival ALMA carbon-monoxide data were used to study the compact dwarf galaxy Pox 186, a local analog of reionization-era systems. Photoionization models favor stripped binary stars as the source of its extreme mid-infrared line ratios, while a young starburst is needed for the dust continuum. The inferred star-formation efficiency exceeds 15 billionths per year and the cold-gas depletion time is short, cautioning against classifying distant galaxies from ultraviolet diagnostics alone. ### Why it matters {#why-it-matters-mp-2026-09-25-014} JWST and ALMA observations point to star formation—not an active galactic nucleus—in Pox 186. ### Limits and context {#limitations-mp-2026-09-25-014} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-09-25-014} - JWST and ALMA observations point to star formation—not an active galactic nucleus—in Pox 186. [source-2026-09-25-014] ## 15. Telemonitoring Flagged Two-Thirds of Heart-Failure Timelines {#mp-2026-09-25-026} - Story ID: `mp-2026-09-25-026` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-026/telemonitoring-flagged-two-thirds-of-heart-failure-timelines **Dek:** TRACER worked from sparse, irregular biomarker measurements in a 276-patient dataset. TRACER used time-aware biomarker embeddings, contrastive pretraining and separate event classifiers on overlapping windows from 276 heart-failure patients. The model correctly identified 66.7% of timelines leading to relevant hospitalization while overestimating 7.9%, according to the authors. Training the task as event detection outperformed direct forecasting in the limited-event setting. The retrospective result suggests an alerting direction; it is not clinical validation or evidence of improved patient outcomes. ### Why it matters {#why-it-matters-mp-2026-09-25-026} TRACER worked from sparse, irregular biomarker measurements in a 276-patient dataset. ### Limits and context {#limitations-mp-2026-09-25-026} - The model correctly identified 66.7% of timelines leading to relevant hospitalization while overestimating 7.9%, according to the authors. - The retrospective result suggests an alerting direction; it is not clinical validation or evidence of improved patient outcomes. ### Claims and sources {#claims-mp-2026-09-25-026} - TRACER worked from sparse, irregular biomarker measurements in a 276-patient dataset. [source-2026-09-25-015] — Qualification: The model correctly identified 66.7% of timelines leading to relevant hospitalization while overestimating 7.9%, according to the authors. ## 16. AI Exploration Sometimes Got Worse With More Searching {#mp-2026-09-25-027} - Story ID: `mp-2026-09-25-027` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-027/ai-exploration-sometimes-got-worse-with-more-searching **Dek:** Two executable alien-world sandboxes separate learning unfamiliar rules from recalling familiar ones. ExplorationBench creates AlienCode and AlienLogic environments whose executable rules conflict with familiar knowledge. Together they contain 55 discovery targets and 140 tasks, each with a flawed manual, environmental feedback and a tool schema. Across ten systems, the strongest could acquire and apply unfamiliar rules, but results varied substantially by trajectory and continued exploration sometimes stalled or reversed earlier gains. The benchmark tests synthetic worlds, not open-ended scientific discovery. ### Why it matters {#why-it-matters-mp-2026-09-25-027} Two executable alien-world sandboxes separate learning unfamiliar rules from recalling familiar ones. ### Limits and context {#limitations-mp-2026-09-25-027} - The benchmark tests synthetic worlds, not open-ended scientific discovery. ### Claims and sources {#claims-mp-2026-09-25-027} - Two executable alien-world sandboxes separate learning unfamiliar rules from recalling familiar ones. [source-2026-09-25-016] — Qualification: The benchmark tests synthetic worlds, not open-ended scientific discovery. ## 17. One Video Became a Reusable Robot Program {#mp-2026-09-25-015} - Story ID: `mp-2026-09-25-015` - Type: `ticker` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-015/one-video-became-a-reusable-robot-program **Dek:** RAPID inferred a task specification, action primitives and a test environment from one human demonstration. The agentic loop generated, executed and refined object-centric robot programs across eight contact-rich tasks and LIBERO-Pro manipulation tasks, then ran all eight nonprehensile tasks on a real Franka arm. The authors report generalization across pose, shape, material and environment. ### Why it matters {#why-it-matters-mp-2026-09-25-015} RAPID inferred a task specification, action primitives and a test environment from one human demonstration. ### Limits and context {#limitations-mp-2026-09-25-015} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-09-25-015} - RAPID inferred a task specification, action primitives and a test environment from one human demonstration. [source-2026-09-25-017] ## 18. A Robot Chose a Support Surface Three Times Faster {#mp-2026-09-25-016} - Story ID: `mp-2026-09-25-016` - Type: `ticker` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-016/a-robot-chose-a-support-surface-three-times-faster **Dek:** CTCS screened reach, mobility and residual-force tradeoffs across 392 task conditions. On a Unitree Go2 quadruped with an AgileX arm and nine support surfaces, selective exact evaluations closely matched the exhaustive objective while producing about a threefold speedup over checking every candidate contact. ### Why it matters {#why-it-matters-mp-2026-09-25-016} CTCS screened reach, mobility and residual-force tradeoffs across 392 task conditions. ### Limits and context {#limitations-mp-2026-09-25-016} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-09-25-016} - CTCS screened reach, mobility and residual-force tradeoffs across 392 task conditions. [source-2026-09-25-018] ## 19. Optocouplers Drove 20 Kilovolts at 500 Hertz {#mp-2026-09-25-017} - Story ID: `mp-2026-09-25-017` - Type: `ticker` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-017/optocouplers-drove-20-kilovolts-at-500-hertz **Dek:** A feedback amplifier treats optocouplers as linear devices for electrostatic actuators. The prototype produced a 20-kilovolt peak-to-peak output with bandwidth up to 500 hertz at sub-milliamp current. The authors report lower distortion and higher bandwidth than equivalent pulse-width-modulated amplifiers with few components. ### Why it matters {#why-it-matters-mp-2026-09-25-017} A feedback amplifier treats optocouplers as linear devices for electrostatic actuators. ### Limits and context {#limitations-mp-2026-09-25-017} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-09-25-017} - A feedback amplifier treats optocouplers as linear devices for electrostatic actuators. [source-2026-09-25-019] ## 20. A Detour Heuristic Ran Four Orders Faster {#mp-2026-09-25-018} - Story ID: `mp-2026-09-25-018` - Type: `ticker` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-018/a-detour-heuristic-ran-four-orders-faster **Dek:** FOLD optimized short-lived entanglement resources instead of accumulating pair rates over time. On tractable one-shot quantum-routing instances, FOLD recovered almost all optimal fidelity and ran four orders of magnitude faster than exact schedule enumeration at circuit rank five. It also led polynomial baselines on tested Internet topologies where routing could help. ### Why it matters {#why-it-matters-mp-2026-09-25-018} FOLD optimized short-lived entanglement resources instead of accumulating pair rates over time. ### Limits and context {#limitations-mp-2026-09-25-018} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-09-25-018} - FOLD optimized short-lived entanglement resources instead of accumulating pair rates over time. [source-2026-09-25-020] ## 21. Sparse Rubin Data Still Yielded Preliminary Asteroid Profiles {#mp-2026-09-25-019} - Story ID: `mp-2026-09-25-019` - Type: `ticker` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-019/sparse-rubin-data-still-yielded-preliminary-asteroid-profiles **Dek:** Multi-band fitting estimated rotation, color and taxonomy from early science-validation observations. The method combines Fourier light-curve fitting and multi-band period analysis, then derives rotation-corrected colors and taxonomic classes. The authors say preliminary properties are possible from sparse irregular data, while robust determinations need better temporal coverage. ### Why it matters {#why-it-matters-mp-2026-09-25-019} Multi-band fitting estimated rotation, color and taxonomy from early science-validation observations. ### Limits and context {#limitations-mp-2026-09-25-019} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-09-25-019} - Multi-band fitting estimated rotation, color and taxonomy from early science-validation observations. [source-2026-09-25-021] ## 22. MicroGroove {#mp-2026-09-25-020} - Story ID: `mp-2026-09-25-020` - Type: `invention_desk` - Classification: `editorial` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-020/microgroove **Dek:** Turns a Cardputer-ADV, a printable shell, and open firmware into a pocket four-track instrument with synthesis, drums, microphone sampling, resampling, and step sequencing. Turns a Cardputer-ADV, a printable shell, and open firmware into a pocket four-track instrument with synthesis, drums, microphone sampling, resampling, and step sequencing. ### Why it matters {#why-it-matters-mp-2026-09-25-020} An independent builder is turning an improbable idea into a working project. ### Limits and context {#limitations-mp-2026-09-25-020} - A Desk Pick is an editorial selection, not a product endorsement. ### Claims and sources {#claims-mp-2026-09-25-020} - This Invention Desk entry makes no independently sourced news claim. ## 23. Sense Cane {#mp-2026-09-25-021} - Story ID: `mp-2026-09-25-021` - Type: `invention_desk` - Classification: `editorial` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-021/sense-cane **Dek:** Combines three ultrasonic modules, a small controller, and one vibration motor so a buildable cane prototype can signal obstacles at different heights without audio, an app, or a phone. Combines three ultrasonic modules, a small controller, and one vibration motor so a buildable cane prototype can signal obstacles at different heights without audio, an app, or a phone. ### Why it matters {#why-it-matters-mp-2026-09-25-021} An independent builder is turning an improbable idea into a working project. ### Limits and context {#limitations-mp-2026-09-25-021} - A Desk Pick is an editorial selection, not a product endorsement. ### Claims and sources {#claims-mp-2026-09-25-021} - This Invention Desk entry makes no independently sourced news claim. ## 24. DIYraman {#mp-2026-09-25-022} - Story ID: `mp-2026-09-25-022` - Type: `invention_desk` - Classification: `editorial` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-022/diyraman **Dek:** Pairs a surplus spectrometer, filtered 532-nanometer excitation, and printable mechanics in a documented Raman setup for optics education and cautious exploratory materials analysis. Pairs a surplus spectrometer, filtered 532-nanometer excitation, and printable mechanics in a documented Raman setup for optics education and cautious exploratory materials analysis. ### Why it matters {#why-it-matters-mp-2026-09-25-022} An independent builder is turning an improbable idea into a working project. ### Limits and context {#limitations-mp-2026-09-25-022} - A Desk Pick is an editorial selection, not a product endorsement. ### Claims and sources {#claims-mp-2026-09-25-022} - This Invention Desk entry makes no independently sourced news claim. ## 25. Neato D10 Brain Transplant {#mp-2026-09-25-023} - Story ID: `mp-2026-09-25-023` - Type: `invention_desk` - Classification: `editorial` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-023/neato-d10-brain-transplant **Dek:** Documents replacing a cloud-disabled robot vacuum's locked control electronics with a Raspberry Pi, an ESP32, and ROS 2 while reusing its chassis, motors, battery, sensors, and lidar. Documents replacing a cloud-disabled robot vacuum's locked control electronics with a Raspberry Pi, an ESP32, and ROS 2 while reusing its chassis, motors, battery, sensors, and lidar. ### Why it matters {#why-it-matters-mp-2026-09-25-023} An independent builder is turning an improbable idea into a working project. ### Limits and context {#limitations-mp-2026-09-25-023} - A Desk Pick is an editorial selection, not a product endorsement. ### Claims and sources {#claims-mp-2026-09-25-023} - This Invention Desk entry makes no independently sourced news claim. ## 26. The First Paid Slot {#mp-2026-09-25-024} - Story ID: `mp-2026-09-25-024` - Type: `invention_desk` - Classification: `house_example` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-024/the-first-paid-slot **Dek:** A transparent preview of paid placement with one verified link and no claim of endorsement. A transparent preview of paid placement with one verified link and no claim of endorsement. House example - no advertiser paid. Payment will buy placement, never endorsement. ### Why it matters {#why-it-matters-mp-2026-09-25-024} This placement explains how builders can appear in The Invention Desk without purchasing editorial endorsement. ### Limits and context {#limitations-mp-2026-09-25-024} - House example - no advertiser paid. Payment will buy placement, never endorsement. ### Claims and sources {#claims-mp-2026-09-25-024} - This Invention Desk entry makes no independently sourced news claim. ## 27. Put Your Project on the Desk {#mp-2026-09-25-025} - Story ID: `mp-2026-09-25-025` - Type: `invention_desk` - Classification: `house_example` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-09-25-025/put-your-project-on-the-desk **Dek:** One manually reviewed placement stays active for seven days and remains separate from Desk Picks. One manually reviewed placement stays active for seven days and remains separate from Desk Picks. Manual intake only. Payment buys placement, never endorsement, and every submission is reviewed. ### Why it matters {#why-it-matters-mp-2026-09-25-025} This placement explains how builders can appear in The Invention Desk without purchasing editorial endorsement. ### Limits and context {#limitations-mp-2026-09-25-025} - Manual intake only. Payment buys placement, never endorsement, and every submission is reviewed. ### Claims and sources {#claims-mp-2026-09-25-025} - This Invention Desk entry makes no independently sourced news claim. ## Normalized sources - **source-2026-09-25-001:** [arXiv preprint 2609.29736](https://arxiv.org/abs/2609.29736) — arXiv; primary_research - **source-2026-09-25-002:** [arXiv preprint 2609.30137](https://arxiv.org/abs/2609.30137) — arXiv; primary_research - **source-2026-09-25-003:** [arXiv preprint 2609.30094](https://arxiv.org/abs/2609.30094) — arXiv; primary_research - **source-2026-09-25-004:** [arXiv preprint 2609.30051](https://arxiv.org/abs/2609.30051) — arXiv; primary_research - **source-2026-09-25-005:** [arXiv preprint 2609.30027](https://arxiv.org/abs/2609.30027) — arXiv; primary_research - **source-2026-09-25-006:** [arXiv preprint 2609.29852](https://arxiv.org/abs/2609.29852) — arXiv; primary_research - **source-2026-09-25-007:** [arXiv preprint 2609.29921](https://arxiv.org/abs/2609.29921) — arXiv; primary_research - **source-2026-09-25-008:** [arXiv preprint 2609.30082](https://arxiv.org/abs/2609.30082) — arXiv; primary_research - **source-2026-09-25-009:** [arXiv preprint 2609.30157](https://arxiv.org/abs/2609.30157) — arXiv; primary_research - **source-2026-09-25-010:** [arXiv preprint 2609.29750](https://arxiv.org/abs/2609.29750) — arXiv; primary_research - **source-2026-09-25-011:** [arXiv preprint 2609.29929](https://arxiv.org/abs/2609.29929) — arXiv; primary_research - **source-2026-09-25-012:** [arXiv preprint 2609.29890](https://arxiv.org/abs/2609.29890) — arXiv; primary_research - **source-2026-09-25-013:** [arXiv preprint 2609.29850](https://arxiv.org/abs/2609.29850) — arXiv; primary_research - **source-2026-09-25-014:** [arXiv preprint 2609.30003](https://arxiv.org/abs/2609.30003) — arXiv; primary_research - **source-2026-09-25-015:** [arXiv preprint 2609.29742](https://arxiv.org/abs/2609.29742) — arXiv; primary_research - **source-2026-09-25-016:** [arXiv preprint 2609.30199](https://arxiv.org/abs/2609.30199) — arXiv; primary_research - **source-2026-09-25-017:** [arXiv preprint 2609.30249](https://arxiv.org/abs/2609.30249) — arXiv; primary_research - **source-2026-09-25-018:** [arXiv preprint 2609.30140](https://arxiv.org/abs/2609.30140) — arXiv; primary_research - **source-2026-09-25-019:** [arXiv preprint 2609.29914](https://arxiv.org/abs/2609.29914) — arXiv; primary_research - **source-2026-09-25-020:** [arXiv preprint 2609.29752](https://arxiv.org/abs/2609.29752) — arXiv; primary_research - **source-2026-09-25-021:** [arXiv preprint 2609.29841](https://arxiv.org/abs/2609.29841) — arXiv; primary_research