--- schema_version: "1.0.0" edition_id: "mp-2026-11-06-morning-0119" published_at: "2026-11-06T09:00:00.000-05:00" modified_at: "2026-11-06T09:00:00.000-05:00" canonical_url: "https://themachinepress.com/edition/2026-11-06" story_count: 27 lead_story_id: "mp-2026-11-06-001" --- # The Machine Press — Morning edition Edition ID: `mp-2026-11-06-morning-0119` Published: 2026-11-06T09:00:00.000-05:00 Canonical edition: https://themachinepress.com/edition/2026-11-06 Boron nitride nanotubes host multiple room-temperature quantum emitters along one compact structure, with surface coating reducing blinking. ## 1. One Nanotube Carries a Spectrum of Single Photons {#mp-2026-11-06-001} - Story ID: `mp-2026-11-06-001` - Type: `lead` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-001/one-nanotube-carries-a-spectrum-of-single-photons **Dek:** Boron nitride nanotubes host multiple room-temperature quantum emitters along one compact structure, with surface coating reducing blinking. Single-photon sources must be bright, stable and small enough to integrate with photonic circuitry. The authors used confocal photoluminescence microscopy and Hanbury Brown–Twiss interferometry to study boron nitride nanotubes at room temperature. They report emitters across the visible spectrum, including regions not previously reported for flat hexagonal boron nitride, with multiple emission sites at different wavelengths along individual tubes. The fastest measured lifetime was about 0.5 nanoseconds, and surface passivation reduced blinking. The result is a laboratory characterization of candidate emitters—not a finished communications device—but a roughly 25-nanometer tube offers a compact geometry for deterministic placement in future waveguides and filters. ### Why it matters {#why-it-matters-mp-2026-11-06-001} Boron nitride nanotubes host multiple room-temperature quantum emitters along one compact structure, with surface coating reducing blinking. ### Limits and context {#limitations-mp-2026-11-06-001} - They report emitters across the visible spectrum, including regions not previously reported for flat hexagonal boron nitride, with multiple emission sites at different wavelengths along individual tubes. - The result is a laboratory characterization of candidate emitters—not a finished communications device—but a roughly 25-nanometer tube offers a compact geometry for deterministic placement in future waveguides and filters. ### Claims and sources {#claims-mp-2026-11-06-001} - Boron nitride nanotubes host multiple room-temperature quantum emitters along one compact structure, with surface coating reducing blinking. [source-2026-11-06-001] — Qualification: They report emitters across the visible spectrum, including regions not previously reported for flat hexagonal boron nitride, with multiple emission sites at different wavelengths along individual tubes. ## 2. A Thousand Papers Become Places to Experiment {#mp-2026-11-06-002} - Story ID: `mp-2026-11-06-002` - Type: `secondary` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-002/a-thousand-papers-become-places-to-experiment **Dek:** Scientist World converts research papers into verified executable environments, pairing reproduced baselines with agent research traces. Scientific papers are readable records, but they rarely arrive as standardized environments in which a research agent can change a method and measure the result. Scientist World uses an autonomous pipeline to extract an open research question from each paper and reproduce the paper’s method as a baseline. The authors release the first 1,000 verified environments plus 6,400 hours of recorded research traces. On their evaluation, frontier research agents surpassed the reproduced human baseline in 74 percent of environments, while training on the traces raised a smaller model’s ideation score by 158 percent to match the frontier systems tested. Those percentages depend on the benchmark’s reproduction and scoring choices; they do not mean agents surpassed scientists generally. The release matters because it turns a claim about automated research into a large, inspectable experimental substrate. ### Why it matters {#why-it-matters-mp-2026-11-06-002} Scientist World converts research papers into verified executable environments, pairing reproduced baselines with agent research traces. ### Limits and context {#limitations-mp-2026-11-06-002} - Those percentages depend on the benchmark’s reproduction and scoring choices; they do not mean agents surpassed scientists generally. ### Claims and sources {#claims-mp-2026-11-06-002} - Scientist World converts research papers into verified executable environments, pairing reproduced baselines with agent research traces. [source-2026-11-06-002] — Qualification: Those percentages depend on the benchmark’s reproduction and scoring choices; they do not mean agents surpassed scientists generally. ## 3. The Swarm Stops Writing Down Every Thought {#mp-2026-11-06-003} - Story ID: `mp-2026-11-06-003` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-003/the-swarm-stops-writing-down-every-thought **Dek:** Hundreds of low-cost agents exchange latent KV caches and select novel hidden-state directions for evolutionary program search. SwarmEvolve organizes agents into groups that reason without decoding every intermediate step to text. A novelty filter measures distances between hidden states, while a summarizer receives combined KV caches to perform crossover. Across 21 mathematical-discovery and scientific-computing problems, the authors report the highest average score and a median 3 to 8 minutes to reach scores other frameworks needed 60 minutes to match. The comparison is limited to the tested problems, models and budgets. ### Why it matters {#why-it-matters-mp-2026-11-06-003} Hundreds of low-cost agents exchange latent KV caches and select novel hidden-state directions for evolutionary program search. ### Limits and context {#limitations-mp-2026-11-06-003} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-11-06-003} - Hundreds of low-cost agents exchange latent KV caches and select novel hidden-state directions for evolutionary program search. [source-2026-11-06-003] ## 4. Repository Migration Exposes the Agent Gap {#mp-2026-11-06-004} - Story ID: `mp-2026-11-06-004` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-004/repository-migration-exposes-the-agent-gap **Dek:** XStack asks coding agents to move complete PyTorch projects on NVIDIA GPUs into JAX repositories running on Google TPUs. Each task begins with a PyTorch repository and a clean TPU machine; a migration passes only when it reproduces authenticated criteria derived from the source project’s published result. In the released benchmark, the strongest reported agent passes 75 percent, GPT-5.6-sol passes 45 percent and every other tested agent reaches at most 40 percent. The scores describe this benchmark’s repositories, infrastructure and grading—not all framework migration work. ### Why it matters {#why-it-matters-mp-2026-11-06-004} XStack asks coding agents to move complete PyTorch projects on NVIDIA GPUs into JAX repositories running on Google TPUs. ### Limits and context {#limitations-mp-2026-11-06-004} - Each task begins with a PyTorch repository and a clean TPU machine; a migration passes only when it reproduces authenticated criteria derived from the source project’s published result. - The scores describe this benchmark’s repositories, infrastructure and grading—not all framework migration work. ### Claims and sources {#claims-mp-2026-11-06-004} - XStack asks coding agents to move complete PyTorch projects on NVIDIA GPUs into JAX repositories running on Google TPUs. [source-2026-11-06-004] — Qualification: Each task begins with a PyTorch repository and a clean TPU machine; a migration passes only when it reproduces authenticated criteria derived from the source project’s published result. ## 5. A Few Corrections Teach the Reason for Failure {#mp-2026-11-06-005} - Story ID: `mp-2026-11-06-005` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-005/a-few-corrections-teach-the-reason-for-failure **Dek:** OopsieRecover infers semantic intent from corrective demonstrations, then steers robot policies with spatial costs across new scenes. Instead of binding a correction to the exact observation in which it appeared, the framework uses abduction to infer why the original behavior was undesirable. It encodes that knowledge as dense spatial representations and conditional costs. Across six manipulation tasks and variations in objects, poses, scenes and related tasks, the authors report that a handful of demonstrations can transfer corrections without fresh data for every variation. The result remains confined to the paper’s tasks and vision-language-action models. ### Why it matters {#why-it-matters-mp-2026-11-06-005} OopsieRecover infers semantic intent from corrective demonstrations, then steers robot policies with spatial costs across new scenes. ### Limits and context {#limitations-mp-2026-11-06-005} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-11-06-005} - OopsieRecover infers semantic intent from corrective demonstrations, then steers robot policies with spatial costs across new scenes. [source-2026-11-06-005] ## 6. The Visual Plan Arrives in One Primitive {#mp-2026-11-06-006} - Story ID: `mp-2026-11-06-006` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-006/the-visual-plan-arrives-in-one-primitive **Dek:** VFlash predicts complete points, boxes or trajectories in one forward pass instead of emitting their coordinates token by token. A lightweight drafter exploits the known structure of each visual primitive and adapts to its type, dimensionality and length. Across the evaluated 2D and 3D robot-trajectory tasks, the method reduces prediction error relative to autoregressive decoding. Preliminary single-rollout measurements show roughly 25-times faster 3D point-motion generation, while a conditioned world-action model reaches 70.2 percent average success across four MolmoSpaces suites versus 62.6 percent for the baseline. ### Why it matters {#why-it-matters-mp-2026-11-06-006} VFlash predicts complete points, boxes or trajectories in one forward pass instead of emitting their coordinates token by token. ### Limits and context {#limitations-mp-2026-11-06-006} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-11-06-006} - VFlash predicts complete points, boxes or trajectories in one forward pass instead of emitting their coordinates token by token. [source-2026-11-06-006] ## 7. Thermal Loss Opens a Classical Route Through Sampling {#mp-2026-11-06-007} - Story ID: `mp-2026-11-06-007` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-007/thermal-loss-opens-a-classical-route-through-sampling **Dek:** A common simulation framework places noisy boson and fermion sampling in polynomial time under fixed nonzero overall loss. The result covers Fock-state boson sampling, fixed-squeezing Gaussian boson sampling and fermion sampling with magic inputs, at inverse-polynomial total-variation error and with high probability over Haar-random interferometers. Under a weaker flatness assumption, the authors obtain quasipolynomial simulation without Haar randomness and polynomial simulation at logarithmic depth with constant loss per layer. These are asymptotic theoretical conditions, not a demonstration that every noisy quantum experiment is classically easy. ### Why it matters {#why-it-matters-mp-2026-11-06-007} A common simulation framework places noisy boson and fermion sampling in polynomial time under fixed nonzero overall loss. ### Limits and context {#limitations-mp-2026-11-06-007} - These are asymptotic theoretical conditions, not a demonstration that every noisy quantum experiment is classically easy. ### Claims and sources {#claims-mp-2026-11-06-007} - A common simulation framework places noisy boson and fermion sampling in polynomial time under fixed nonzero overall loss. [source-2026-11-06-007] — Qualification: These are asymptotic theoretical conditions, not a demonstration that every noisy quantum experiment is classically easy. ## 8. The Robot Monitor Warns Before the Failure {#mp-2026-11-06-008} - Story ID: `mp-2026-11-06-008` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-008/the-robot-monitor-warns-before-the-failure **Dek:** VLA-Tug probes latent layers for calibrated early failure detection and classifies the likely failure mode with little overhead. Training on the pre-failure segment improves early detection by as much as 30 percent at matched false-positive rates over the strongest learned baseline. Under unseen-task shift, missed-detection and false-alarm bounds remain valid in up to 86 percent of runs, versus at most 21 percent for baselines. Top-three failure-mode accuracy rises by up to 44 percent over a majority baseline, and the authors release 3,180 annotated manipulation episodes. These are benchmarked monitoring results, not a safety certification. ### Why it matters {#why-it-matters-mp-2026-11-06-008} VLA-Tug probes latent layers for calibrated early failure detection and classifies the likely failure mode with little overhead. ### Limits and context {#limitations-mp-2026-11-06-008} - Under unseen-task shift, missed-detection and false-alarm bounds remain valid in up to 86 percent of runs, versus at most 21 percent for baselines. - These are benchmarked monitoring results, not a safety certification. ### Claims and sources {#claims-mp-2026-11-06-008} - VLA-Tug probes latent layers for calibrated early failure detection and classifies the likely failure mode with little overhead. [source-2026-11-06-008] — Qualification: Under unseen-task shift, missed-detection and false-alarm bounds remain valid in up to 86 percent of runs, versus at most 21 percent for baselines. ## 9. Thirty Percent of the Benchmark Needed Repair {#mp-2026-11-06-009} - Story ID: `mp-2026-11-06-009` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-009/thirty-percent-of-the-benchmark-needed-repair **Dek:** A continuous-validation audit turns Terminal-Bench 2.0 into 2.1 and shows that task defects and time budgets both move results. The workflow combines community reports, task artifacts, traces, automated oracle checks and human-gated repairs. The audit finds issues in roughly 30 percent of Terminal-Bench 2.0; repairs raise accuracy by more than 10 percentage points and move leaderboard positions by up to three ranks, though rank correlation remains high. Extending wall-clock limits also closes large harness gaps for two tested models, showing that resource budgets belong in the benchmark specification. ### Why it matters {#why-it-matters-mp-2026-11-06-009} A continuous-validation audit turns Terminal-Bench 2.0 into 2.1 and shows that task defects and time budgets both move results. ### Limits and context {#limitations-mp-2026-11-06-009} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-11-06-009} - A continuous-validation audit turns Terminal-Bench 2.0 into 2.1 and shows that task defects and time budgets both move results. [source-2026-11-06-009] ## 10. The Humanoid Learns to Move With a Partner {#mp-2026-11-06-010} - Story ID: `mp-2026-11-06-010` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-010/the-humanoid-learns-to-move-with-a-partner **Dek:** A hierarchical controller conditions whole-body robot motion on a human partner and anticipates motion through temporary occlusion. Social Humanoid uses a shared tokenized representation for human and robot movement, then a low-level controller executes the generated motion on a 29-degree-of-freedom Unitree G1. Adaptive retargeting across five motion-capture datasets yields a 260-hour interaction benchmark. The authors report coordinated behavior across varied interaction types in simulation and real-world demonstrations. The evidence concerns the released system and scenarios, not open-ended social understanding. ### Why it matters {#why-it-matters-mp-2026-11-06-010} A hierarchical controller conditions whole-body robot motion on a human partner and anticipates motion through temporary occlusion. ### Limits and context {#limitations-mp-2026-11-06-010} - The evidence concerns the released system and scenarios, not open-ended social understanding. ### Claims and sources {#claims-mp-2026-11-06-010} - A hierarchical controller conditions whole-body robot motion on a human partner and anticipates motion through temporary occlusion. [source-2026-11-06-010] — Qualification: The evidence concerns the released system and scenarios, not open-ended social understanding. ## 11. Reasoning Restarts at the First Suspected Error {#mp-2026-11-06-011} - Story ID: `mp-2026-11-06-011` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-011/reasoning-restarts-at-the-first-suspected-error **Dek:** StepBack localizes a likely first mistake, then resamples only that step and the suffix rather than whole reasoning traces. The method begins with a small set of draft trajectories, uses lightweight signals to find likely first-error positions and votes across corrected suffixes. Across seven math-reasoning and multi-hop question-answering benchmarks, the authors report average accuracy matching self-consistency while using up to 41.26 percent fewer tokens. It can also be combined with adaptive-consistency methods. The savings depend on the tested tasks and localization signals. ### Why it matters {#why-it-matters-mp-2026-11-06-011} StepBack localizes a likely first mistake, then resamples only that step and the suffix rather than whole reasoning traces. ### Limits and context {#limitations-mp-2026-11-06-011} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-11-06-011} - StepBack localizes a likely first mistake, then resamples only that step and the suffix rather than whole reasoning traces. [source-2026-11-06-011] ## 12. One Surface Changes Both Warmth and Softness {#mp-2026-11-06-012} - Story ID: `mp-2026-11-06-012` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-012/one-surface-changes-both-warmth-and-softness **Dek:** A particle-jamming and water-transfer prototype independently modulates stiffness and temperature for tactile displays. Mechanical tests show a controllable 300 percent stiffness increase. Thermal characterization spans 10 to 50 degrees Celsius, with closed-loop control maintaining a 0.2-degree error from its setpoint. The prototype is a research display for perceptual work, medical simulation and affective haptics; the paper does not establish durability, wearability or clinical performance. ### Why it matters {#why-it-matters-mp-2026-11-06-012} A particle-jamming and water-transfer prototype independently modulates stiffness and temperature for tactile displays. ### Limits and context {#limitations-mp-2026-11-06-012} - The prototype is a research display for perceptual work, medical simulation and affective haptics; the paper does not establish durability, wearability or clinical performance. ### Claims and sources {#claims-mp-2026-11-06-012} - A particle-jamming and water-transfer prototype independently modulates stiffness and temperature for tactile displays. [source-2026-11-06-012] — Qualification: The prototype is a research display for perceptual work, medical simulation and affective haptics; the paper does not establish durability, wearability or clinical performance. ## 13. The Agents Send Only the Latent Pieces That Still Matter {#mp-2026-11-06-013} - Story ID: `mp-2026-11-06-013` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-013/the-agents-send-only-the-latent-pieces-that-still-matter **Dek:** SCALE reuses similar tokens across time and schedules semantic gain against bandwidth and real-time deadlines. The method matches tokens between consecutive latent messages, profiles the gain and cost of sending different amounts of representation, and schedules pending transmissions under bandwidth and deadline limits. Across three cyber-physical benchmarks and a physical spider-robot case study, the authors report higher task performance and on-time delivery in constrained networks. The results do not establish general network efficiency outside the tested models and conditions. ### Why it matters {#why-it-matters-mp-2026-11-06-013} SCALE reuses similar tokens across time and schedules semantic gain against bandwidth and real-time deadlines. ### Limits and context {#limitations-mp-2026-11-06-013} - The results do not establish general network efficiency outside the tested models and conditions. ### Claims and sources {#claims-mp-2026-11-06-013} - SCALE reuses similar tokens across time and schedules semantic gain against bandwidth and real-time deadlines. [source-2026-11-06-013] — Qualification: The results do not establish general network efficiency outside the tested models and conditions. ## 14. The Sensor Reading Skips Untrusted Memory {#mp-2026-11-06-014} - Story ID: `mp-2026-11-06-014` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-014/the-sensor-reading-skips-untrusted-memory **Dek:** SensorGuard-RV authenticates samples in fixed-function hardware and decrypts them directly into a RISC-V register. A write-once ownership switch hands the sensor to a controller whose key is inaccessible to software. A new instruction verifies tags and rejects replays, including across resets, before placing data in a CPU register. On an FPGA implementation with an accelerometer, a protected sample arrives in 51.9 microseconds and the loop sustains 19,015 transactions per second, 5.3 percent above the software-driven SPI baseline. Added logic and estimated power remain material tradeoffs. ### Why it matters {#why-it-matters-mp-2026-11-06-014} SensorGuard-RV authenticates samples in fixed-function hardware and decrypts them directly into a RISC-V register. ### Limits and context {#limitations-mp-2026-11-06-014} - Added logic and estimated power remain material tradeoffs. ### Claims and sources {#claims-mp-2026-11-06-014} - SensorGuard-RV authenticates samples in fixed-function hardware and decrypts them directly into a RISC-V register. [source-2026-11-06-014] — Qualification: Added logic and estimated power remain material tradeoffs. ## 15. JWST Measures the Most Distant Known Ultra-Cold Y Dwarf {#mp-2026-11-06-026} - Story ID: `mp-2026-11-06-026` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-026/jwst-measures-the-most-distant-known-ultra-cold-y-dwarf **Dek:** A two-year NIRCam baseline places JADES-GS-BD-5 about 64 parsecs away, roughly three times farther than prior parallax-measured Y dwarfs. The team measures 49.8 milliarcseconds per year of proper motion and a parallax of 15.7 milliarcseconds, yielding a geometric distance of 64 parsecs with stated uncertainty of plus six and minus five. Its inferred motion fits the Galactic thin disk, while its faint luminosity and near-infrared spectrum place it beyond Y1 and among the coldest dwarfs known. Evolution models put it in the planetary-mass regime, though current atmosphere models do not reproduce the full spectral-energy distribution. ### Why it matters {#why-it-matters-mp-2026-11-06-026} A two-year NIRCam baseline places JADES-GS-BD-5 about 64 parsecs away, roughly three times farther than prior parallax-measured Y dwarfs. ### Limits and context {#limitations-mp-2026-11-06-026} - Evolution models put it in the planetary-mass regime, though current atmosphere models do not reproduce the full spectral-energy distribution. ### Claims and sources {#claims-mp-2026-11-06-026} - A two-year NIRCam baseline places JADES-GS-BD-5 about 64 parsecs away, roughly three times farther than prior parallax-measured Y dwarfs. [source-2026-11-06-015] — Qualification: Evolution models put it in the planetary-mass regime, though current atmosphere models do not reproduce the full spectral-energy distribution. ## 16. The Failure Predictor Mostly Recognizes the Issue and Model {#mp-2026-11-06-027} - Story ID: `mp-2026-11-06-027` - Type: `dispatch` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-027/the-failure-predictor-mostly-recognizes-the-issue-and-model **Dek:** On public coding-agent trajectories, issue difficulty and model identity explain far more than the partial run itself. A run-independent issue-difficulty score reaches 0.92 AUROC on SWE-bench Verified trajectories, while trajectory predictors reach about 0.62. Within an issue, model identity predicts failure better than the studied predictors, and per-issue model switching provides no held-out gain. A repeated-run pilot shows that retries can rescue some failures, but current predictors cannot distinguish a bad run from a task beyond the model’s ability. The result argues for repeated-run, closed-loop evaluation rather than universal pessimism about monitoring. ### Why it matters {#why-it-matters-mp-2026-11-06-027} On public coding-agent trajectories, issue difficulty and model identity explain far more than the partial run itself. ### Limits and context {#limitations-mp-2026-11-06-027} - A repeated-run pilot shows that retries can rescue some failures, but current predictors cannot distinguish a bad run from a task beyond the model’s ability. ### Claims and sources {#claims-mp-2026-11-06-027} - On public coding-agent trajectories, issue difficulty and model identity explain far more than the partial run itself. [source-2026-11-06-016] — Qualification: A repeated-run pilot shows that retries can rescue some failures, but current predictors cannot distinguish a bad run from a task beyond the model’s ability. ## 17. Compression Targets the Hardware, Not the Parameter Count {#mp-2026-11-06-015} - Story ID: `mp-2026-11-06-015` - Type: `ticker` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-015/compression-targets-the-hardware-not-the-parameter-count **Dek:** AutoCompress searches architectures against latency, throughput and memory models validated on actual serving stacks. The authors report higher average accuracy than Minitron across RTX 5070, RTX 5080 and H100 settings, and compress Nemotron 3.5 Super from 120 billion to 74 billion parameters for one B200 at 128K context while retaining 95 percent of its mean benchmark score. ### Why it matters {#why-it-matters-mp-2026-11-06-015} AutoCompress searches architectures against latency, throughput and memory models validated on actual serving stacks. ### Limits and context {#limitations-mp-2026-11-06-015} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-11-06-015} - AutoCompress searches architectures against latency, throughput and memory models validated on actual serving stacks. [source-2026-11-06-017] ## 18. The Visual Assistant Enters a Closed Loop {#mp-2026-11-06-016} - Story ID: `mp-2026-11-06-016` - Type: `ticker` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-016/the-visual-assistant-enters-a-closed-loop **Dek:** IVG-Bench simulates users and executable tasks across cooking, household manipulation and mobile interfaces. Across 315 scenarios, the best tested model finishes 53.4 percent within the time limit. A human study finds 78.5 percent agreement with the interaction-quality metric, and a small-model training experiment improves task success and quality in the cooking environment. ### Why it matters {#why-it-matters-mp-2026-11-06-016} IVG-Bench simulates users and executable tasks across cooking, household manipulation and mobile interfaces. ### Limits and context {#limitations-mp-2026-11-06-016} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-11-06-016} - IVG-Bench simulates users and executable tasks across cooking, household manipulation and mobile interfaces. [source-2026-11-06-018] ## 19. The Repair Runs Cleanly and Returns the Wrong Answer {#mp-2026-11-06-017} - Story ID: `mp-2026-11-06-017` - Type: `ticker` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-017/the-repair-runs-cleanly-and-returns-the-wrong-answer **Dek:** A preregistered study finds typical-input gates release 41 percent of wrong repairs after runtime or API migrations. One edge input derived from the release note cuts that rate to 6.5 percent across 73 tasks. Hard distractors remain difficult, with 15.3 to 16.8 percent of wrong programs passing two stronger frozen checks. ### Why it matters {#why-it-matters-mp-2026-11-06-017} A preregistered study finds typical-input gates release 41 percent of wrong repairs after runtime or API migrations. ### Limits and context {#limitations-mp-2026-11-06-017} - Hard distractors remain difficult, with 15.3 to 16.8 percent of wrong programs passing two stronger frozen checks. ### Claims and sources {#claims-mp-2026-11-06-017} - A preregistered study finds typical-input gates release 41 percent of wrong repairs after runtime or API migrations. [source-2026-11-06-019] — Qualification: Hard distractors remain difficult, with 15.3 to 16.8 percent of wrong programs passing two stronger frozen checks. ## 20. Quantum Error-Correction Distances Get Exact Certificates {#mp-2026-11-06-018} - Story ID: `mp-2026-11-06-018` - Type: `ticker` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-018/quantum-error-correction-distances-get-exact-certificates **Dek:** Rank-based dynamic programming certifies codes up to 31,200 qubits and circuits with 800,000 fault mechanisms. An audit of 26 recent papers finds overestimated or non-tight distances in ten. For one IBM memory circuit, the certified distance is 16 rather than the assumed 18; redesigned schedules recover 18 with an extra timestep. ### Why it matters {#why-it-matters-mp-2026-11-06-018} Rank-based dynamic programming certifies codes up to 31,200 qubits and circuits with 800,000 fault mechanisms. ### Limits and context {#limitations-mp-2026-11-06-018} - No additional limitation was separately recorded. ### Claims and sources {#claims-mp-2026-11-06-018} - Rank-based dynamic programming certifies codes up to 31,200 qubits and circuits with 800,000 fault mechanisms. [source-2026-11-06-020] ## 21. Quantum Software Stacks Get Differential Tests {#mp-2026-11-06-019} - Story ID: `mp-2026-11-06-019` - Type: `ticker` - Classification: `editorial` - Content status: `new` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-019/quantum-software-stacks-get-differential-tests **Dek:** QDiff generates semantics-preserving program variants, filters costly circuits and compares probabilistic outputs. The framework uses static circuit properties to avoid uninformative hardware runs and supports distribution comparisons including Kolmogorov–Smirnov tests and cross entropy. The paper proposes and evaluates a testing method; it does not certify any whole quantum stack as defect-free. ### Why it matters {#why-it-matters-mp-2026-11-06-019} QDiff generates semantics-preserving program variants, filters costly circuits and compares probabilistic outputs. ### Limits and context {#limitations-mp-2026-11-06-019} - The paper proposes and evaluates a testing method; it does not certify any whole quantum stack as defect-free. ### Claims and sources {#claims-mp-2026-11-06-019} - QDiff generates semantics-preserving program variants, filters costly circuits and compares probabilistic outputs. [source-2026-11-06-021] — Qualification: The paper proposes and evaluates a testing method; it does not certify any whole quantum stack as defect-free. ## 22. Space Drums 2.0 {#mp-2026-11-06-020} - Story ID: `mp-2026-11-06-020` - Type: `invention_desk` - Classification: `editorial` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-020/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. 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 {#why-it-matters-mp-2026-11-06-020} An independent builder is turning an improbable idea into a working project. ### Limits and context {#limitations-mp-2026-11-06-020} - A Desk Pick is an editorial selection, not a product endorsement. ### Claims and sources {#claims-mp-2026-11-06-020} - This Invention Desk entry makes no independently sourced news claim. ## 23. 3RIC {#mp-2026-11-06-021} - Story ID: `mp-2026-11-06-021` - Type: `invention_desk` - Classification: `editorial` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-021/3ric **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. 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 {#why-it-matters-mp-2026-11-06-021} An independent builder is turning an improbable idea into a working project. ### Limits and context {#limitations-mp-2026-11-06-021} - A Desk Pick is an editorial selection, not a product endorsement. ### Claims and sources {#claims-mp-2026-11-06-021} - This Invention Desk entry makes no independently sourced news claim. ## 24. M3-CRETE {#mp-2026-11-06-022} - Story ID: `mp-2026-11-06-022` - Type: `invention_desk` - Classification: `editorial` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-022/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. 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 {#why-it-matters-mp-2026-11-06-022} An independent builder is turning an improbable idea into a working project. ### Limits and context {#limitations-mp-2026-11-06-022} - A Desk Pick is an editorial selection, not a product endorsement. ### Claims and sources {#claims-mp-2026-11-06-022} - This Invention Desk entry makes no independently sourced news claim. ## 25. DeskNode {#mp-2026-11-06-023} - Story ID: `mp-2026-11-06-023` - Type: `invention_desk` - Classification: `editorial` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-023/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. 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 {#why-it-matters-mp-2026-11-06-023} An independent builder is turning an improbable idea into a working project. ### Limits and context {#limitations-mp-2026-11-06-023} - A Desk Pick is an editorial selection, not a product endorsement. ### Claims and sources {#claims-mp-2026-11-06-023} - This Invention Desk entry makes no independently sourced news claim. ## 26. The First Paid Slot {#mp-2026-11-06-024} - Story ID: `mp-2026-11-06-024` - Type: `invention_desk` - Classification: `house_example` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-024/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-2026-11-06-024} This placement explains how builders can appear in The Invention Desk without purchasing editorial endorsement. ### Limits and context {#limitations-mp-2026-11-06-024} - House example - no advertiser paid. Payment will buy placement, never endorsement. ### Claims and sources {#claims-mp-2026-11-06-024} - This Invention Desk entry makes no independently sourced news claim. ## 27. Put Your Project on the Desk {#mp-2026-11-06-025} - Story ID: `mp-2026-11-06-025` - Type: `invention_desk` - Classification: `house_example` - Content status: `carried_over` - Permanent URL: https://themachinepress.com/story/mp-2026-11-06-025/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-2026-11-06-025} This placement explains how builders can appear in The Invention Desk without purchasing editorial endorsement. ### Limits and context {#limitations-mp-2026-11-06-025} - Manual intake only. Payment buys placement, never endorsement, and every submission is reviewed. ### Claims and sources {#claims-mp-2026-11-06-025} - This Invention Desk entry makes no independently sourced news claim. ## Normalized sources - **source-2026-11-06-001:** [arXiv preprint 2611.08460, original v1](https://arxiv.org/abs/2611.08460v1) — arXiv; primary_research - **source-2026-11-06-002:** [arXiv preprint 2611.08478, original v1](https://arxiv.org/abs/2611.08478v1) — arXiv; primary_research - **source-2026-11-06-003:** [arXiv preprint 2611.08483, original v1](https://arxiv.org/abs/2611.08483v1) — arXiv; primary_research - **source-2026-11-06-004:** [arXiv preprint 2611.08472, original v1](https://arxiv.org/abs/2611.08472v1) — arXiv; primary_research - **source-2026-11-06-005:** [arXiv preprint 2611.08462, original v1](https://arxiv.org/abs/2611.08462v1) — arXiv; primary_research - **source-2026-11-06-006:** [arXiv preprint 2611.08449, original v1](https://arxiv.org/abs/2611.08449v1) — arXiv; primary_research - **source-2026-11-06-007:** [arXiv preprint 2611.08444, original v1](https://arxiv.org/abs/2611.08444v1) — arXiv; primary_research - **source-2026-11-06-008:** [arXiv preprint 2611.08425, original v1](https://arxiv.org/abs/2611.08425v1) — arXiv; primary_research - **source-2026-11-06-009:** [arXiv preprint 2611.08417, original v1](https://arxiv.org/abs/2611.08417v1) — arXiv; primary_research - **source-2026-11-06-010:** [arXiv preprint 2611.08407, original v1](https://arxiv.org/abs/2611.08407v1) — arXiv; primary_research - **source-2026-11-06-011:** [arXiv preprint 2611.08403, original v1](https://arxiv.org/abs/2611.08403v1) — arXiv; primary_research - **source-2026-11-06-012:** [arXiv preprint 2611.08398, original v1](https://arxiv.org/abs/2611.08398v1) — arXiv; primary_research - **source-2026-11-06-013:** [arXiv preprint 2611.08390, original v1](https://arxiv.org/abs/2611.08390v1) — arXiv; primary_research - **source-2026-11-06-014:** [arXiv preprint 2611.08380, original v1](https://arxiv.org/abs/2611.08380v1) — arXiv; primary_research - **source-2026-11-06-015:** [arXiv preprint 2611.08346, original v1](https://arxiv.org/abs/2611.08346v1) — arXiv; primary_research - **source-2026-11-06-016:** [arXiv preprint 2611.08285, original v1](https://arxiv.org/abs/2611.08285v1) — arXiv; primary_research - **source-2026-11-06-017:** [arXiv preprint 2611.08290, original v1](https://arxiv.org/abs/2611.08290v1) — arXiv; primary_research - **source-2026-11-06-018:** [arXiv preprint 2611.08286, original v1](https://arxiv.org/abs/2611.08286v1) — arXiv; primary_research - **source-2026-11-06-019:** [arXiv preprint 2611.08269, original v1](https://arxiv.org/abs/2611.08269v1) — arXiv; primary_research - **source-2026-11-06-020:** [arXiv preprint 2611.08247, original v1](https://arxiv.org/abs/2611.08247v1) — arXiv; primary_research - **source-2026-11-06-021:** [arXiv preprint 2611.08220, original v1](https://arxiv.org/abs/2611.08220v1) — arXiv; primary_research