---
schema_version: "1.0.0"
edition_id: "mp-2026-09-19-morning-0072"
published_at: "2026-09-19T09:00:00.000-04:00"
modified_at: "2026-09-19T09:00:00.000-04:00"
canonical_url: "https://themachinepress.com/edition/2026-09-19"
story_count: 27
lead_story_id: "mp-2026-09-19-001"
---

# The Machine Press — Morning edition

Edition ID: `mp-2026-09-19-morning-0072`  
Published: 2026-09-19T09:00:00.000-04:00  
Canonical edition: https://themachinepress.com/edition/2026-09-19

A geometric gate cut displaced source error by 60%, while one diffusion proposal took 41.7 milliseconds.

## 1. The Robot Followed a Smell Through the Wind {#mp-2026-09-19-001}

- Story ID: `mp-2026-09-19-001`
- Type: `lead`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-001/the-robot-followed-a-smell-through-the-wind

**Dek:** A geometric gate cut displaced source error by 60%, while one diffusion proposal took 41.7 milliseconds.

SmellDiffusion represents species-specific gas zones in an open-vocabulary scene graph, estimates a named leak's source, and gives the same goal to classical and diffusion planners. In 424 solved source-wind configurations, 28 concentration peaks sat more than half a meter from the true source. A gate trained only on the training split detected nine of ten held-out displacements at 0.64 precision and reduced mean error on displaced cases from 1.468 meters to 0.592 meters. One diffusion proposal ran faster than gas-guided A*, but best-of-ten sequential sampling cost more overall; plain A* remained the fastest shortest-path method.

### Why it matters {#why-it-matters-mp-2026-09-19-001}

A geometric gate cut displaced source error by 60%, while one diffusion proposal took 41.7 milliseconds.

### Limits and context {#limitations-mp-2026-09-19-001}

- A gate trained only on the training split detected nine of ten held-out displacements at 0.64 precision and reduced mean error on displaced cases from 1.468 meters to 0.592 meters.

### Claims and sources {#claims-mp-2026-09-19-001}

- A geometric gate cut displaced source error by 60%, while one diffusion proposal took 41.7 milliseconds. [source-2026-09-19-001] — Qualification: A gate trained only on the training split detected nine of ten held-out displacements at 0.64 precision and reduced mean error on displaced cases from 1.468 meters to 0.592 meters.

## 2. Water Clouds Moved Across the Coldest Brown Dwarf {#mp-2026-09-19-002}

- Story ID: `mp-2026-09-19-002`
- Type: `secondary`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-002/water-clouds-moved-across-the-coldest-brown-dwarf

**Dek:** JWST watched WISE 0855 for 11 hours and found carbon-monoxide modulation up to 10% at some wavelengths.

JWST/NIRSpec followed WISE 0855, a 265-kelvin brown dwarf, every 15 minutes across 2.87 to 5.27 micrometers. Carbon monoxide produced the strongest variable spectral feature, reaching a peak-to-peak amplitude of 10% at some wavelengths. Principal-component analysis linked the carbon-monoxide and phosphine variations, while comparisons with atmosphere and structure models supported patchy water clouds. The authors place the chemical variations at deeper quenched pressures and the changing cloud thickness higher in the atmosphere; the result is accepted to The Astronomical Journal but remains an interpretation of spectra, not a direct image of clouds.

### Why it matters {#why-it-matters-mp-2026-09-19-002}

JWST watched WISE 0855 for 11 hours and found carbon-monoxide modulation up to 10% at some wavelengths.

### Limits and context {#limitations-mp-2026-09-19-002}

- The authors place the chemical variations at deeper quenched pressures and the changing cloud thickness higher in the atmosphere; the result is accepted to The Astronomical Journal but remains an interpretation of spectra, not a direct image of clouds.

### Claims and sources {#claims-mp-2026-09-19-002}

- JWST watched WISE 0855 for 11 hours and found carbon-monoxide modulation up to 10% at some wavelengths. [source-2026-09-19-002] — Qualification: The authors place the chemical variations at deeper quenched pressures and the changing cloud thickness higher in the atmosphere; the result is accepted to The Astronomical Journal but remains an interpretation of spectra, not a direct image of clouds.

## 3. The Agent Harness Cut Token Traffic Almost in Half {#mp-2026-09-19-003}

- Story ID: `mp-2026-09-19-003`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-003/the-agent-harness-cut-token-traffic-almost-in-half

**Dek:** SoL-Pi matched its comparison harness while reducing recorded token traffic 44.7% to 49.0%.

The SoL-Pi team scaled automated harness experiments across many environments, then retained four mechanisms covering action execution, context compaction, observation handling and delegated reading. On the 51-task EdgeBench evaluation, the resulting harness performed comparably to Pi across GPT-5.6 Sol and Opus 5 while cutting recorded token traffic by 44.7% to 49.0% and reported API cost by about one third. The savings are evaluation estimates from the tested models, prices and tasks, not a universal operating-cost guarantee.

### Why it matters {#why-it-matters-mp-2026-09-19-003}

SoL-Pi matched its comparison harness while reducing recorded token traffic 44.7% to 49.0%.

### Limits and context {#limitations-mp-2026-09-19-003}

- The savings are evaluation estimates from the tested models, prices and tasks, not a universal operating-cost guarantee.

### Claims and sources {#claims-mp-2026-09-19-003}

- SoL-Pi matched its comparison harness while reducing recorded token traffic 44.7% to 49.0%. [source-2026-09-19-003] — Qualification: The savings are evaluation estimates from the tested models, prices and tasks, not a universal operating-cost guarantee.

## 4. The Agent Groups Agreed—Mostly on the Wrong Answer {#mp-2026-09-19-004}

- Story ID: `mp-2026-09-19-004`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-004/the-agent-groups-agreed-mostly-on-the-wrong-answer

**Dek:** Matched language-model groups exceeded human consensus by 34.0 to 44.4 percentage points in sensitivity analyses.

Researchers replayed 100 held-out human Wason-task groups with language-model agents anchored to each participant's initial answer. Human consensus estimates moved from 24% to 57% depending on who counted as participating, yet two sensitivity analyses still found model groups 34.0 to 44.4 percentage points more consensual. Reasoning-mode agents agreed nearly unanimously after a change designed to remove a memorizable answer, but usually agreed on the wrong answer. In this setting, synthetic agreement did not estimate human deliberation or collective accuracy.

### Why it matters {#why-it-matters-mp-2026-09-19-004}

Matched language-model groups exceeded human consensus by 34.0 to 44.4 percentage points in sensitivity analyses.

### Limits and context {#limitations-mp-2026-09-19-004}

- In this setting, synthetic agreement did not estimate human deliberation or collective accuracy.

### Claims and sources {#claims-mp-2026-09-19-004}

- Matched language-model groups exceeded human consensus by 34.0 to 44.4 percentage points in sensitivity analyses. [source-2026-09-19-004] — Qualification: In this setting, synthetic agreement did not estimate human deliberation or collective accuracy.

## 5. Attribution Could Not Reliably Filter Subliminal Learning {#mp-2026-09-19-005}

- Story ID: `mp-2026-09-19-005`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-005/attribution-could-not-reliably-filter-subliminal-learning

**Dek:** One gradient method mitigated part of the effect at token level, but results changed across models and preferences.

Subliminal learning can transmit behavioral traits through training examples that do not state those traits, limiting semantic filters. The study tested GradCos, a contrastive variant and EK-FAC against divergence tokens across three models. Token-level EK-FAC removed a meaningful part of the effect, while the other attribution methods offered little benefit and generally trailed the counterfactual-teacher baseline. Whole-sample filtering was weaker for every method, and no approach worked consistently across model-preference combinations.

### Why it matters {#why-it-matters-mp-2026-09-19-005}

One gradient method mitigated part of the effect at token level, but results changed across models and preferences.

### Limits and context {#limitations-mp-2026-09-19-005}

- Subliminal learning can transmit behavioral traits through training examples that do not state those traits, limiting semantic filters.

### Claims and sources {#claims-mp-2026-09-19-005}

- One gradient method mitigated part of the effect at token level, but results changed across models and preferences. [source-2026-09-19-005] — Qualification: Subliminal learning can transmit behavioral traits through training examples that do not state those traits, limiting semantic filters.

## 6. Troubleshooting Retrieval Started in the Middle of the Case {#mp-2026-09-19-006}

- Story ID: `mp-2026-09-19-006`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-006/troubleshooting-retrieval-started-in-the-middle-of-the-case

**Dek:** RAFT matched the live incident to an intermediate historical state, then returned the surrounding resolution path.

Static retrieval can find a similar closed ticket without recognizing which stage of its history resembles the active incident. RAFT converts each resolved case into a chain of timeline entries, retrieves at entry level, and returns the parent trajectory around the matched state; an optional graph links similar cases. It beat vanilla RAG and GraphRAG on a synthetic Windows Server benchmark at every stage, with statistically significant gains over the strongest baseline. Apache Jira duplicate labels supplied directional real-case evidence, not a production-agent deployment test.

### Why it matters {#why-it-matters-mp-2026-09-19-006}

RAFT matched the live incident to an intermediate historical state, then returned the surrounding resolution path.

### Limits and context {#limitations-mp-2026-09-19-006}

- Apache Jira duplicate labels supplied directional real-case evidence, not a production-agent deployment test.

### Claims and sources {#claims-mp-2026-09-19-006}

- RAFT matched the live incident to an intermediate historical state, then returned the surrounding resolution path. [source-2026-09-19-006] — Qualification: Apache Jira duplicate labels supplied directional real-case evidence, not a production-agent deployment test.

## 7. The Robot Learned When One Subtask Was Actually Over {#mp-2026-09-19-007}

- Story ID: `mp-2026-09-19-007`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-007/the-robot-learned-when-one-subtask-was-actually-over

**Dek:** A small vision-language model distilled teacher explanations into real-time stage-transition decisions.

Long-horizon robot systems must decide when to stop one skill and start the next, but hand-built completion checkers are brittle and large cloud models are slow. StageGuard combines teacher-model reasoning with demonstrations to generate explanations of policy switching, then trains a lightweight vision-language model to emit compact self-explanations and transition decisions. Evaluation covered two trajectory benchmarks, BEHAVIOR-1K closed-loop control and real robots. The abstract reports substantial prediction gains without enough figures to support a broader quantitative claim.

### Why it matters {#why-it-matters-mp-2026-09-19-007}

A small vision-language model distilled teacher explanations into real-time stage-transition decisions.

### Limits and context {#limitations-mp-2026-09-19-007}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-19-007}

- A small vision-language model distilled teacher explanations into real-time stage-transition decisions. [source-2026-09-19-007]

## 8. Driving Policies Improved With Three Orders Fewer Simulator Interactions {#mp-2026-09-19-008}

- Story ID: `mp-2026-09-19-008`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-008/driving-policies-improved-with-three-orders-fewer-simulator-interactions

**Dek:** A vector-input teacher lifted two camera-policy driving scores by 1.6× and 9.5× in AlpaSim.

OPTED separates reinforcement learning from camera-policy post-training. A privileged teacher learns from vectorized maps and boxes, then supervises pretrained TransFuser and VaVAM students in closed loop on neural reconstructions of real driving logs. The reported driving scores increased 1.6 times and 9.5 times. In controlled experiments, the approach matched direct reinforcement-learning post-training with roughly one-thousandth as many simulator interactions while remaining closer to the human demonstration prior; this is simulation evidence, not public-road validation.

### Why it matters {#why-it-matters-mp-2026-09-19-008}

A vector-input teacher lifted two camera-policy driving scores by 1.6× and 9.5× in AlpaSim.

### Limits and context {#limitations-mp-2026-09-19-008}

- In controlled experiments, the approach matched direct reinforcement-learning post-training with roughly one-thousandth as many simulator interactions while remaining closer to the human demonstration prior; this is simulation evidence, not public-road validation.

### Claims and sources {#claims-mp-2026-09-19-008}

- A vector-input teacher lifted two camera-policy driving scores by 1.6× and 9.5× in AlpaSim. [source-2026-09-19-008] — Qualification: In controlled experiments, the approach matched direct reinforcement-learning post-training with roughly one-thousandth as many simulator interactions while remaining closer to the human demonstration prior; this is simulation evidence, not public-road validation.

## 9. The Underwater Tracker Split Yaw From Translation {#mp-2026-09-19-009}

- Story ID: `mp-2026-09-19-009`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-009/the-underwater-tracker-split-yaw-from-translation

**Dek:** A target-specific depth mask and two motion models kept the vehicle's control problem small enough for real time.

The underwater tracking system selects target depth pixels using color, disparity and temporal cues, then filters depth and image-center estimates separately. Its controller handles yaw apart from translation, avoiding a larger coupled optimization, while predictive control blends constant-velocity and stationary-target models according to recent errors. Actuation, following distance and field of view remain explicit constraints. Simulations and physical experiments beat the authors' comparison frameworks, though the abstract does not quantify that margin.

### Why it matters {#why-it-matters-mp-2026-09-19-009}

A target-specific depth mask and two motion models kept the vehicle's control problem small enough for real time.

### Limits and context {#limitations-mp-2026-09-19-009}

- Its controller handles yaw apart from translation, avoiding a larger coupled optimization, while predictive control blends constant-velocity and stationary-target models according to recent errors.
- Actuation, following distance and field of view remain explicit constraints.
- Simulations and physical experiments beat the authors' comparison frameworks, though the abstract does not quantify that margin.

### Claims and sources {#claims-mp-2026-09-19-009}

- A target-specific depth mask and two motion models kept the vehicle's control problem small enough for real time. [source-2026-09-19-009] — Qualification: Its controller handles yaw apart from translation, avoiding a larger coupled optimization, while predictive control blends constant-velocity and stationary-target models according to recent errors.

## 10. One Mission Plan Crossed From Air to Water {#mp-2026-09-19-010}

- Story ID: `mp-2026-09-19-010`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-010/one-mission-plan-crossed-from-air-to-water

**Dek:** A PX4 extension added marine modes and mixed air-water waypoints without removing the flight safety stack.

Hybrid drones can fly to an aquatic site and then navigate on the surface, but the mission software must change modes without splitting the plan. This PX4 extension adds manual and autonomous marine navigation, keeps existing flight functions and safety mechanisms, and exposes distinct aerial and marine waypoints through the standard mission pipeline and QGroundControl. Simulated missions produced stable surface navigation in calm and wavy conditions. The paper validates firmware behavior in simulation, not a completed field campaign.

### Why it matters {#why-it-matters-mp-2026-09-19-010}

A PX4 extension added marine modes and mixed air-water waypoints without removing the flight safety stack.

### Limits and context {#limitations-mp-2026-09-19-010}

- The paper validates firmware behavior in simulation, not a completed field campaign.

### Claims and sources {#claims-mp-2026-09-19-010}

- A PX4 extension added marine modes and mixed air-water waypoints without removing the flight safety stack. [source-2026-09-19-010] — Qualification: The paper validates firmware behavior in simulation, not a completed field campaign.

## 11. The Sea Urchin Detector Trained in a Synthetic Ocean {#mp-2026-09-19-011}

- Story ID: `mp-2026-09-19-011`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-011/the-sea-urchin-detector-trained-in-a-synthetic-ocean

**Dek:** OceanSim generated labeled underwater scenes with configurable appearance, structure and sensors.

Labeled underwater imagery is costly, so the team extended an IsaacSim-based OceanSim environment with an automatic synthetic-data pipeline. It can vary scene appearance, structure and sensor settings while producing photorealistic images and labels. A real-world sea-urchin detector served as the sim-to-real test, with experiments probing how kinds of scene variation affect transfer. The authors release code and emphasize remaining limits in rendering fidelity, diversity and generalization rather than claiming that synthetic data replaces field collection.

### Why it matters {#why-it-matters-mp-2026-09-19-011}

OceanSim generated labeled underwater scenes with configurable appearance, structure and sensors.

### Limits and context {#limitations-mp-2026-09-19-011}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-19-011}

- OceanSim generated labeled underwater scenes with configurable appearance, structure and sensors. [source-2026-09-19-011]

## 12. Animal Gaits Learned to Move Sideways and Backward {#mp-2026-09-19-012}

- Story ID: `mp-2026-09-19-012`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-012/animal-gaits-learned-to-move-sideways-and-backward

**Dek:** Dynamics-completed demonstrations cut quadruped velocity-tracking error 63.1% on a Cartesian command grid.

Animal demonstrations usually show only a few directions. OmniMimic combines time reversal, constrained dynamics completion and sagittal reflection to create robot-specific supervision beyond those motions, then widens commands toward target velocity ranges. A shared policy uses gait-specialized residual experts for corrections. Across four simulated gaits, it reduced foot-position error 12.9% at forward and backward reference speeds and velocity-tracking error 63.1% on a uniform command grid versus the matched baseline.

### Why it matters {#why-it-matters-mp-2026-09-19-012}

Dynamics-completed demonstrations cut quadruped velocity-tracking error 63.1% on a Cartesian command grid.

### Limits and context {#limitations-mp-2026-09-19-012}

- Animal demonstrations usually show only a few directions.

### Claims and sources {#claims-mp-2026-09-19-012}

- Dynamics-completed demonstrations cut quadruped velocity-tracking error 63.1% on a Cartesian command grid. [source-2026-09-19-012] — Qualification: Animal demonstrations usually show only a few directions.

## 13. One Control Field Cut Qubit Leakage by Three Orders {#mp-2026-09-19-013}

- Story ID: `mp-2026-09-19-013`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-013/one-control-field-cut-qubit-leakage-by-three-orders

**Dek:** An analytic pulse correction suppressed simulated fast-iSWAP leakage without a second control channel.

Leakage-suppression recipes such as DRAG commonly assume two orthogonal controls, but baseband flux architectures may provide only one real-valued field. A Magnus-expansion construction shapes that single field for a tunable-coupler two-transmon gate. With realistic simulated parameters, the correction reduced non-adiabatic leakage from the coupler ramp by as much as three orders of magnitude. The result is a general analytic control proposal supported by device simulation, not an experimental gate demonstration.

### Why it matters {#why-it-matters-mp-2026-09-19-013}

An analytic pulse correction suppressed simulated fast-iSWAP leakage without a second control channel.

### Limits and context {#limitations-mp-2026-09-19-013}

- Leakage-suppression recipes such as DRAG commonly assume two orthogonal controls, but baseband flux architectures may provide only one real-valued field.
- The result is a general analytic control proposal supported by device simulation, not an experimental gate demonstration.

### Claims and sources {#claims-mp-2026-09-19-013}

- An analytic pulse correction suppressed simulated fast-iSWAP leakage without a second control channel. [source-2026-09-19-013] — Qualification: Leakage-suppression recipes such as DRAG commonly assume two orthogonal controls, but baseband flux architectures may provide only one real-valued field.

## 14. Blind Quantum Computing Shrink-Wrapped the Server {#mp-2026-09-19-014}

- Story ID: `mp-2026-09-19-014`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-014/blind-quantum-computing-shrink-wrapped-the-server

**Dek:** The server size now scales with non-Clifford gates rather than the full delegated computation.

Blind quantum computation lets a low-resource client hide a delegated computation, but conventional server resources grow with the whole circuit. By using Pauli-based computation, this protocol makes server size depend only on the non-Clifford gate count. It inherits fault tolerance and qubit virtualization from that model and adds a dual entanglement-based protocol for resource-state computation with lower execution cost. The work is a protocol result, not a hardware deployment.

### Why it matters {#why-it-matters-mp-2026-09-19-014}

The server size now scales with non-Clifford gates rather than the full delegated computation.

### Limits and context {#limitations-mp-2026-09-19-014}

- By using Pauli-based computation, this protocol makes server size depend only on the non-Clifford gate count.
- The work is a protocol result, not a hardware deployment.

### Claims and sources {#claims-mp-2026-09-19-014}

- The server size now scales with non-Clifford gates rather than the full delegated computation. [source-2026-09-19-014] — Qualification: By using Pauli-based computation, this protocol makes server size depend only on the non-Clifford gate count.

## 15. The Phase Detector Wrote Its Decision as Spin Correlators {#mp-2026-09-19-026}

- Story ID: `mp-2026-09-19-026`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-026/the-phase-detector-wrote-its-decision-as-spin-correlators

**Dek:** Tetris-shaped filters turned noisy simulator snapshots into symbolic, experimentally measurable formulas.

Neural networks can locate phase transitions but rarely explain their order parameters. TetrisCNN runs differently shaped convolutional filters in parallel and learns sparse latent variables expressed as spin correlators. On experimental snapshots from two-dimensional Ising and XY quantum simulators measured in several bases, it detected transitions and crossovers while turning its representation and decision boundary into symbolic formulas. The claim is interpretability on the tested systems, not automatic discovery of every phase of matter.

### Why it matters {#why-it-matters-mp-2026-09-19-026}

Tetris-shaped filters turned noisy simulator snapshots into symbolic, experimentally measurable formulas.

### Limits and context {#limitations-mp-2026-09-19-026}

- The claim is interpretability on the tested systems, not automatic discovery of every phase of matter.

### Claims and sources {#claims-mp-2026-09-19-026}

- Tetris-shaped filters turned noisy simulator snapshots into symbolic, experimentally measurable formulas. [source-2026-09-19-015] — Qualification: The claim is interpretability on the tested systems, not automatic discovery of every phase of matter.

## 16. The Wrong Atmosphere Could Hide a Biased Neutron-Star Radius {#mp-2026-09-19-027}

- Story ID: `mp-2026-09-19-027`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-027/the-wrong-atmosphere-could-hide-a-biased-neutron-star-radius

**Dek:** At 20× stronger spot contrast, hydrogen-versus-helium mismatch shifted radius estimates while leaving acceptable residuals.

Synthetic NICER pulse data modeled on the roughly 2.1-solar-mass pulsar PSR J0740+6620 showed little radius bias from choosing hydrogen instead of helium, or vice versa, at the observed spot-to-background ratio. When the ratio increased about twentyfold with total counts held near 550,000, the wrong atmosphere produced significant radius bias while still fitting with statistically acceptable residuals. Bayesian evidence consistently preferred the correct composition, making model comparison—not residual inspection alone—the safeguard.

### Why it matters {#why-it-matters-mp-2026-09-19-027}

At 20× stronger spot contrast, hydrogen-versus-helium mismatch shifted radius estimates while leaving acceptable residuals.

### Limits and context {#limitations-mp-2026-09-19-027}

- Bayesian evidence consistently preferred the correct composition, making model comparison—not residual inspection alone—the safeguard.

### Claims and sources {#claims-mp-2026-09-19-027}

- At 20× stronger spot contrast, hydrogen-versus-helium mismatch shifted radius estimates while leaving acceptable residuals. [source-2026-09-19-016] — Qualification: Bayesian evidence consistently preferred the correct composition, making model comparison—not residual inspection alone—the safeguard.

## 17. Long Gamma-Ray Bursts Doubled Their Supernova-Less Candidate List {#mp-2026-09-19-015}

- Story ID: `mp-2026-09-19-015`
- Type: `ticker`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-015/long-gamma-ray-bursts-doubled-their-supernova-less-candidate-list

**Dek:** Two decades of Swift data produced a rate of 0.5 ± 0.2 events per cubic gigaparsec per year before beaming correction.

A search below redshift 0.35 more than doubled the candidate set with robust supernova limits and distance scales. The small, heterogeneous sample spans star-forming, dusty and quiescent hosts; plausible beaming corrections remain compatible with several compact-merger channels but do not establish one origin.

### Why it matters {#why-it-matters-mp-2026-09-19-015}

Two decades of Swift data produced a rate of 0.5 ± 0.2 events per cubic gigaparsec per year before beaming correction.

### Limits and context {#limitations-mp-2026-09-19-015}

- The small, heterogeneous sample spans star-forming, dusty and quiescent hosts; plausible beaming corrections remain compatible with several compact-merger channels but do not establish one origin.

### Claims and sources {#claims-mp-2026-09-19-015}

- Two decades of Swift data produced a rate of 0.5 ± 0.2 events per cubic gigaparsec per year before beaming correction. [source-2026-09-19-017] — Qualification: The small, heterogeneous sample spans star-forming, dusty and quiescent hosts; plausible beaming corrections remain compatible with several compact-merger channels but do not establish one origin.

## 18. Mixed Fiber Cores Sorted H-Band Light Passively {#mp-2026-09-19-016}

- Story ID: `mp-2026-09-19-016`
- Type: `ticker`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-016/mixed-fiber-cores-sorted-h-band-light-passively

**Dek:** The same 1×7 lanterns collected visible light broadly and separated 1520–1610-nanometer bands by spatial route.

Heterogeneous bundles of SM450 and SMF28 cores break modal degeneracy without a rigid geometry. Experiments found broadband collection from 350 to 740 nanometers and passive spectral sorting in the H band, a potential route to feeding band-optimized astronomical spectrographs.

### Why it matters {#why-it-matters-mp-2026-09-19-016}

The same 1×7 lanterns collected visible light broadly and separated 1520–1610-nanometer bands by spatial route.

### Limits and context {#limitations-mp-2026-09-19-016}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-19-016}

- The same 1×7 lanterns collected visible light broadly and separated 1520–1610-nanometer bands by spatial route. [source-2026-09-19-018]

## 19. Real-Time Burst Search Agreed With Offline Processing {#mp-2026-09-19-017}

- Story ID: `mp-2026-09-19-017`
- Type: `ticker`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-017/real-time-burst-search-agreed-with-offline-processing

**Dek:** Heimdall's ring-buffer and file modes stayed within 3% on arrival time and dispersion measure, and within 10% on signal-to-noise.

Tests on Northern Cross telescope data compared PSRDADA memory buffers with SIGPROC files. Real-time execution kept pace for most runs containing true fast radio bursts, while the authors frame the work as preparatory for beamformer integration rather than a finished observatory pipeline.

### Why it matters {#why-it-matters-mp-2026-09-19-017}

Heimdall's ring-buffer and file modes stayed within 3% on arrival time and dispersion measure, and within 10% on signal-to-noise.

### Limits and context {#limitations-mp-2026-09-19-017}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-19-017}

- Heimdall's ring-buffer and file modes stayed within 3% on arrival time and dispersion measure, and within 10% on signal-to-noise. [source-2026-09-19-019]

## 20. A Nonlinear Amplifier Preserved Every Phase Angle {#mp-2026-09-19-018}

- Story ID: `mp-2026-09-19-018`
- Type: `ticker`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-018/a-nonlinear-amplifier-preserved-every-phase-angle

**Dek:** The constrained noise-limited class is optimized for Yurke–Stoler cat states and Kerr kitten states.

The theory characterizes deterministic nonlinear amplifiers that treat all angles of bosonic phase space uniformly while remaining at the quantum noise limit. Their restricted form supplies an optimum for two families of nonclassical states; the paper establishes a mathematical class, not an experimental device.

### Why it matters {#why-it-matters-mp-2026-09-19-018}

The constrained noise-limited class is optimized for Yurke–Stoler cat states and Kerr kitten states.

### Limits and context {#limitations-mp-2026-09-19-018}

- Their restricted form supplies an optimum for two families of nonclassical states; the paper establishes a mathematical class, not an experimental device.

### Claims and sources {#claims-mp-2026-09-19-018}

- The constrained noise-limited class is optimized for Yurke–Stoler cat states and Kerr kitten states. [source-2026-09-19-020] — Qualification: Their restricted form supplies an optimum for two families of nonclassical states; the paper establishes a mathematical class, not an experimental device.

## 21. Twenty Frames Removed Nearly One-Fifth of Hidden-Vehicle Seeds {#mp-2026-09-19-019}

- Story ID: `mp-2026-09-19-019`
- Type: `ticker`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-019/twenty-frames-removed-nearly-one-fifth-of-hidden-vehicle-seeds

**Dek:** History-conditioned minimax search found six avoidable counterexamples across eight Waymo scenarios.

The method certifies hidden trajectories against visibility, occupancy, map and kinematic constraints before searching for the legal path that hurts the ego vehicle's best response most. Extending memory from one frame to twenty reduced retained hidden seeds about 18.45% overall; no collision-producing attacker was found in two scenes within the finite budget.

### Why it matters {#why-it-matters-mp-2026-09-19-019}

History-conditioned minimax search found six avoidable counterexamples across eight Waymo scenarios.

### Limits and context {#limitations-mp-2026-09-19-019}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-19-019}

- History-conditioned minimax search found six avoidable counterexamples across eight Waymo scenarios. [source-2026-09-19-021]

## 22. SATURNIX {#mp-2026-09-19-020}

- Story ID: `mp-2026-09-19-020`
- Type: `invention_desk`
- Classification: `editorial`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-020/saturnix

**Dek:** Builds a tactile digital camera around a Raspberry Pi Zero 2 W, an autofocus sensor, a small viewfinder, and mechanical-switch controls while publishing the software and printable hardware files.

Builds a tactile digital camera around a Raspberry Pi Zero 2 W, an autofocus sensor, a small viewfinder, and mechanical-switch controls while publishing the software and printable hardware files.

### Why it matters {#why-it-matters-mp-2026-09-19-020}

An independent builder is turning an improbable idea into a working project.

### Limits and context {#limitations-mp-2026-09-19-020}

- A Desk Pick is an editorial selection, not a product endorsement.

### Claims and sources {#claims-mp-2026-09-19-020}

- This Invention Desk entry makes no independently sourced news claim.

## 23. WURB-2026 {#mp-2026-09-19-021}

- Story ID: `mp-2026-09-19-021`
- Type: `invention_desk`
- Classification: `editorial`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-021/wurb-2026

**Dek:** Combines a Raspberry Pi-class computer, an ultrasonic microphone, local storage, and a web interface into a modular recorder for active or unattended bat monitoring.

Combines a Raspberry Pi-class computer, an ultrasonic microphone, local storage, and a web interface into a modular recorder for active or unattended bat monitoring.

### Why it matters {#why-it-matters-mp-2026-09-19-021}

An independent builder is turning an improbable idea into a working project.

### Limits and context {#limitations-mp-2026-09-19-021}

- A Desk Pick is an editorial selection, not a product endorsement.

### Claims and sources {#claims-mp-2026-09-19-021}

- This Invention Desk entry makes no independently sourced news claim.

## 24. OpenCTD {#mp-2026-09-19-022}

- Story ID: `mp-2026-09-19-022`
- Type: `invention_desk`
- Classification: `editorial`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-022/openctd

**Dek:** Packages conductivity, temperature, and depth sensors with an Arduino-compatible controller, battery, and SD storage inside a user-built housing for nearshore research and education.

Packages conductivity, temperature, and depth sensors with an Arduino-compatible controller, battery, and SD storage inside a user-built housing for nearshore research and education.

### Why it matters {#why-it-matters-mp-2026-09-19-022}

An independent builder is turning an improbable idea into a working project.

### Limits and context {#limitations-mp-2026-09-19-022}

- A Desk Pick is an editorial selection, not a product endorsement.

### Claims and sources {#claims-mp-2026-09-19-022}

- This Invention Desk entry makes no independently sourced news claim.

## 25. Microban {#mp-2026-09-19-023}

- Story ID: `mp-2026-09-19-023`
- Type: `invention_desk`
- Classification: `editorial`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-023/microban

**Dek:** Turns printable body parts, nineteen servomotors, a Raspberry Pi Zero 2 W, batteries, and shared control software into a compact humanoid platform for hands-on robotics work.

Turns printable body parts, nineteen servomotors, a Raspberry Pi Zero 2 W, batteries, and shared control software into a compact humanoid platform for hands-on robotics work.

### Why it matters {#why-it-matters-mp-2026-09-19-023}

An independent builder is turning an improbable idea into a working project.

### Limits and context {#limitations-mp-2026-09-19-023}

- A Desk Pick is an editorial selection, not a product endorsement.

### Claims and sources {#claims-mp-2026-09-19-023}

- This Invention Desk entry makes no independently sourced news claim.

## 26. The First Paid Slot {#mp-2026-09-19-024}

- Story ID: `mp-2026-09-19-024`
- Type: `invention_desk`
- Classification: `house_example`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-024/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-2026-09-19-024}

This placement explains how builders can appear in The Invention Desk without purchasing editorial endorsement.

### Limits and context {#limitations-mp-2026-09-19-024}

- House example - no advertiser paid. Payment will buy placement, never endorsement.

### Claims and sources {#claims-mp-2026-09-19-024}

- This Invention Desk entry makes no independently sourced news claim.

## 27. Put Your Project on the Desk {#mp-2026-09-19-025}

- Story ID: `mp-2026-09-19-025`
- Type: `invention_desk`
- Classification: `house_example`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-19-025/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-2026-09-19-025}

This placement explains how builders can appear in The Invention Desk without purchasing editorial endorsement.

### Limits and context {#limitations-mp-2026-09-19-025}

- Manual intake only. Payment buys placement, never endorsement, and every submission is reviewed.

### Claims and sources {#claims-mp-2026-09-19-025}

- This Invention Desk entry makes no independently sourced news claim.

## Normalized sources

- **source-2026-09-19-001:** [arXiv preprint 2609.20624](https://arxiv.org/abs/2609.20624) — arXiv; primary_research
- **source-2026-09-19-002:** [arXiv preprint 2609.20664](https://arxiv.org/abs/2609.20664) — arXiv; primary_research
- **source-2026-09-19-003:** [arXiv preprint 2609.20519](https://arxiv.org/abs/2609.20519) — arXiv; primary_research
- **source-2026-09-19-004:** [arXiv preprint 2609.20543](https://arxiv.org/abs/2609.20543) — arXiv; primary_research
- **source-2026-09-19-005:** [arXiv preprint 2609.20027](https://arxiv.org/abs/2609.20027) — arXiv; primary_research
- **source-2026-09-19-006:** [arXiv preprint 2609.20754](https://arxiv.org/abs/2609.20754) — arXiv; primary_research
- **source-2026-09-19-007:** [arXiv preprint 2609.20791](https://arxiv.org/abs/2609.20791) — arXiv; primary_research
- **source-2026-09-19-008:** [arXiv preprint 2609.20756](https://arxiv.org/abs/2609.20756) — arXiv; primary_research
- **source-2026-09-19-009:** [arXiv preprint 2609.20731](https://arxiv.org/abs/2609.20731) — arXiv; primary_research
- **source-2026-09-19-010:** [arXiv preprint 2609.20691](https://arxiv.org/abs/2609.20691) — arXiv; primary_research
- **source-2026-09-19-011:** [arXiv preprint 2609.20680](https://arxiv.org/abs/2609.20680) — arXiv; primary_research
- **source-2026-09-19-012:** [arXiv preprint 2609.20566](https://arxiv.org/abs/2609.20566) — arXiv; primary_research
- **source-2026-09-19-013:** [arXiv preprint 2609.20766](https://arxiv.org/abs/2609.20766) — arXiv; primary_research
- **source-2026-09-19-014:** [arXiv preprint 2609.20729](https://arxiv.org/abs/2609.20729) — arXiv; primary_research
- **source-2026-09-19-015:** [arXiv preprint 2609.20693](https://arxiv.org/abs/2609.20693) — arXiv; primary_research
- **source-2026-09-19-016:** [arXiv preprint 2609.20795](https://arxiv.org/abs/2609.20795) — arXiv; primary_research
- **source-2026-09-19-017:** [arXiv preprint 2609.20787](https://arxiv.org/abs/2609.20787) — arXiv; primary_research
- **source-2026-09-19-018:** [arXiv preprint 2609.20642](https://arxiv.org/abs/2609.20642) — arXiv; primary_research
- **source-2026-09-19-019:** [arXiv preprint 2609.20635](https://arxiv.org/abs/2609.20635) — arXiv; primary_research
- **source-2026-09-19-020:** [arXiv preprint 2609.20644](https://arxiv.org/abs/2609.20644) — arXiv; primary_research
- **source-2026-09-19-021:** [arXiv preprint 2609.20480](https://arxiv.org/abs/2609.20480) — arXiv; primary_research