---
schema_version: "1.0.0"
edition_id: "mp-2026-09-29-morning-0082"
published_at: "2026-09-29T09:00:00.000-04:00"
modified_at: "2026-09-29T09:00:00.000-04:00"
canonical_url: "https://themachinepress.com/edition/2026-09-29"
story_count: 27
lead_story_id: "mp-2026-09-29-001"
---

# The Machine Press — Morning edition

Edition ID: `mp-2026-09-29-morning-0082`  
Published: 2026-09-29T09:00:00.000-04:00  
Canonical edition: https://themachinepress.com/edition/2026-09-29

DexRoam kept locomotion, two-arm motion and finger dexterity coupled as demonstrations crossed embodiments.

## 1. Human Motion Doubled a Robot's Success Without Doubling Robot Data {#mp-2026-09-29-001}

- Story ID: `mp-2026-09-29-001`
- Type: `lead`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-001/human-motion-doubled-a-robot-s-success-without-doubling-robot-data

**Dek:** DexRoam kept locomotion, two-arm motion and finger dexterity coupled as demonstrations crossed embodiments.

DexRoam uses a consumer VR headset and head-mounted stereo camera to capture continuous whole-body human manipulation without external trackers. Three alignment stages map embodiment, action meaning and timing into a mobile bimanual robot's action space, allowing human and robot demonstrations to train standard vision-language-action policies together. In the authors' real-world tests, adding human demonstrations raised average success from 29% to 56% with GR00T N1.7 and from 32% to 57% with pi0.5; the system matched robot-only training while using half as many robot demonstrations. Those figures describe the reported tasks and backbones, not a general guarantee for dexterous robots.

### Why it matters {#why-it-matters-mp-2026-09-29-001}

DexRoam kept locomotion, two-arm motion and finger dexterity coupled as demonstrations crossed embodiments.

### Limits and context {#limitations-mp-2026-09-29-001}

- In the authors' real-world tests, adding human demonstrations raised average success from 29% to 56% with GR00T N1.7 and from 32% to 57% with pi0.5; the system matched robot-only training while using half as many robot demonstrations.
- Those figures describe the reported tasks and backbones, not a general guarantee for dexterous robots.

### Claims and sources {#claims-mp-2026-09-29-001}

- DexRoam kept locomotion, two-arm motion and finger dexterity coupled as demonstrations crossed embodiments. [source-2026-09-29-001] — Qualification: In the authors' real-world tests, adding human demonstrations raised average success from 29% to 56% with GR00T N1.7 and from 32% to 57% with pi0.5; the system matched robot-only training while using half as many robot demonstrations.

## 2. A Failed Tool Still Produced a Success Claim {#mp-2026-09-29-002}

- Story ID: `mp-2026-09-29-002`
- Type: `secondary`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-002/a-failed-tool-still-produced-a-success-claim

**Dek:** A structured evidence contract cut false-success reports from 22.8% to 0.8% in a controlled benchmark.

Failure-Transparent Agents fixes the failed observation and required evidence state before a model responds, separating post-failure reporting from tool choice and recovery. The benchmark contains 100 deterministic failure traces across five failure families, a neutral control and four user-pressure conditions. Across six models, three policies and 3,600 human-annotated responses, the authors report false-success rates of 22.8% under a baseline policy, 9.3% with a transparency instruction and 0.8% with a structured evidence contract. Fabricated-detail rates fell from 28.3% to 0.8% across the same endpoints while useful responses rose from 74.9% to 98.8%. This is a controlled blocked-task benchmark; it supports the tested reporting intervention, not a claim about all agents in open environments.

### Why it matters {#why-it-matters-mp-2026-09-29-002}

A structured evidence contract cut false-success reports from 22.8% to 0.8% in a controlled benchmark.

### Limits and context {#limitations-mp-2026-09-29-002}

- This is a controlled blocked-task benchmark; it supports the tested reporting intervention, not a claim about all agents in open environments.

### Claims and sources {#claims-mp-2026-09-29-002}

- A structured evidence contract cut false-success reports from 22.8% to 0.8% in a controlled benchmark. [source-2026-09-29-002] — Qualification: This is a controlled blocked-task benchmark; it supports the tested reporting intervention, not a claim about all agents in open environments.

## 3. Human Hand States Reset Three Different Robot Bodies {#mp-2026-09-29-003}

- Story ID: `mp-2026-09-29-003`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-003/human-hand-states-reset-three-different-robot-bodies

**Dek:** X-Reset used filtered hand-object states for exploration rather than asking policies to imitate human motion.

X-Reset retargets human hand-object states into noisy robot states, removes unstable configurations in simulation and samples the rest as reinforcement-learning resets. The policy itself conditions on object state and goal, while demonstrations enter through the reset distribution. The authors trained generalist policies on 20 objects across a 22-degree-of-freedom hand mounted on two arms and a parallel-jaw gripper, reporting unseen-object generalization and zero-shot sim-to-real transfer. The preprint's results are limited to its objects, embodiments and simulator-to-hardware setup.

### Why it matters {#why-it-matters-mp-2026-09-29-003}

X-Reset used filtered hand-object states for exploration rather than asking policies to imitate human motion.

### Limits and context {#limitations-mp-2026-09-29-003}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-29-003}

- X-Reset used filtered hand-object states for exploration rather than asking policies to imitate human motion. [source-2026-09-29-003]

## 4. One Language Model Ran at Every One of Its Twenty Depths {#mp-2026-09-29-004}

- Story ID: `mp-2026-09-29-004`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-004/one-language-model-ran-at-every-one-of-its-twenty-depths

**Dek:** Stochastic prefix supervision trained a continuum of capacity instead of a few fixed exits.

Telescopic Language Models train one randomly truncated layer prefix alongside one full-capacity pass on every step. On a 200-million-parameter proxy trained over 20 billion FineWeb-Edu tokens, the resulting model remained usable at all 20 layer prefixes. The authors report a 43% to 44% reduction in area under the quality-budget curve versus fixed-exit suites, matching full-capacity quality at about 12% lower GPU cost per run. These are proxy-scale results, and deployment savings will depend on serving hardware and workload.

### Why it matters {#why-it-matters-mp-2026-09-29-004}

Stochastic prefix supervision trained a continuum of capacity instead of a few fixed exits.

### Limits and context {#limitations-mp-2026-09-29-004}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-29-004}

- Stochastic prefix supervision trained a continuum of capacity instead of a few fixed exits. [source-2026-09-29-004]

## 5. The Image Model Learned to Critique Its Own Revisions {#mp-2026-09-29-005}

- Story ID: `mp-2026-09-29-005`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-005/the-image-model-learned-to-critique-its-own-revisions

**Dek:** One trajectory-level advantage updated both reflection tokens and image edits across repeated repair rounds.

UMM-Reflection trains a unified multimodal model across complete diagnose-and-revise trajectories rather than optimizing text reflection or image generation alone. Sibling attempts share an initial image, and one group-relative advantage updates both the reflection tokens and flow-based revisions. On BAGEL, the authors report a 12.05-point GenEval improvement over supervised fine-tuning, with gains on three held-out evaluation suites. The benchmark gains do not establish that every self-critique is accurate or that the method transfers unchanged to other architectures.

### Why it matters {#why-it-matters-mp-2026-09-29-005}

One trajectory-level advantage updated both reflection tokens and image edits across repeated repair rounds.

### Limits and context {#limitations-mp-2026-09-29-005}

- The benchmark gains do not establish that every self-critique is accurate or that the method transfers unchanged to other architectures.

### Claims and sources {#claims-mp-2026-09-29-005}

- One trajectory-level advantage updated both reflection tokens and image edits across repeated repair rounds. [source-2026-09-29-005] — Qualification: The benchmark gains do not establish that every self-critique is accurate or that the method transfers unchanged to other architectures.

## 6. Finance Agents Got Rubrics With Values Frozen to a Cutoff {#mp-2026-09-29-006}

- Story ID: `mp-2026-09-29-006`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-006/finance-agents-got-rubrics-with-values-frozen-to-a-cutoff

**Dek:** Expert guidance governed a writer, reviewer and code checks that generated task-specific scoring criteria.

FinAutoRubric turns reusable expert guidance into query-specific rubrics for financial research agents. A writer researches expected values, a reviewer checks them, code enforces rules and unresolved failures escalate to a human. Across three expert-authored finance benchmarks, the authors report scoring agreement comparable with their strongest generator and a blind analyst preference for the generated rubrics. The released benchmark covers 100 queries across 78 tasks and eight asset classes; it evaluates rubric production, not investment performance.

### Why it matters {#why-it-matters-mp-2026-09-29-006}

Expert guidance governed a writer, reviewer and code checks that generated task-specific scoring criteria.

### Limits and context {#limitations-mp-2026-09-29-006}

- The released benchmark covers 100 queries across 78 tasks and eight asset classes; it evaluates rubric production, not investment performance.

### Claims and sources {#claims-mp-2026-09-29-006}

- Expert guidance governed a writer, reviewer and code checks that generated task-specific scoring criteria. [source-2026-09-29-006] — Qualification: The released benchmark covers 100 queries across 78 tasks and eight asset classes; it evaluates rubric production, not investment performance.

## 7. One Robot Graph Compiled Into Control and Simulation {#mp-2026-09-29-007}

- Story ID: `mp-2026-09-29-007`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-007/one-robot-graph-compiled-into-control-and-simulation

**Dek:** RoboCompiler kept closure, actuation and dynamics consistent across several looped mechanisms.

RoboCompiler starts from a canonical mechanism graph and produces closure paths, analytic residual Jacobians, feasible configurations, velocity maps and projected dynamics. The team evaluated excavator, quadruped, robot-arm, humanoid and Stewart-platform mechanisms, with independent checks and execution in MuJoCo and Isaac Sim. For the Kangaroo platform, the paper reports a 96.7% reduction in residual-and-Jacobian evaluation time and a 66.8% reduction in closed-loop rollout wall time with dynamics and control held fixed. Those gains are implementation-specific, not universal compiler speedups.

### Why it matters {#why-it-matters-mp-2026-09-29-007}

RoboCompiler kept closure, actuation and dynamics consistent across several looped mechanisms.

### Limits and context {#limitations-mp-2026-09-29-007}

- Those gains are implementation-specific, not universal compiler speedups.

### Claims and sources {#claims-mp-2026-09-29-007}

- RoboCompiler kept closure, actuation and dynamics consistent across several looped mechanisms. [source-2026-09-29-007] — Qualification: Those gains are implementation-specific, not universal compiler speedups.

## 8. An Agent's Token Budget Was Forecast While It Ran {#mp-2026-09-29-008}

- Story ID: `mp-2026-09-29-008`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-008/an-agent-s-token-budget-was-forecast-while-it-ran

**Dek:** TokenCast composed segment costs and context growth without making another model call.

TokenCast records each execution segment's own token use and the context growth it adds, then composes segments to estimate repeated input costs later in a run. On SWE-bench Verified, the authors report a mean cumulative prediction time of 32.8 milliseconds per run. Across four task suites and six agent models, mean absolute error improved by an average 14.5% over the strongest comparator; an offline replay used 21.3% fewer tokens than a fixed-budget policy at matched trace completion. Replay results do not prove identical savings in live production systems.

### Why it matters {#why-it-matters-mp-2026-09-29-008}

TokenCast composed segment costs and context growth without making another model call.

### Limits and context {#limitations-mp-2026-09-29-008}

- Replay results do not prove identical savings in live production systems.

### Claims and sources {#claims-mp-2026-09-29-008}

- TokenCast composed segment costs and context growth without making another model call. [source-2026-09-29-008] — Qualification: Replay results do not prove identical savings in live production systems.

## 9. A Humanoid Model Split Action Tokens by Body Part {#mp-2026-09-29-009}

- Story ID: `mp-2026-09-29-009`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-009/a-humanoid-model-split-action-tokens-by-body-part

**Dek:** Grouped discrete diffusion decoded end-effector, body, hand and kinematic actions for real-time control.

Holo-M extends a language model's vocabulary with four action-token groups for end effectors, body, hands and kinematics. The design trains across humanoid teleoperation, egocentric human video and simulation, then decodes body-part groups with discrete diffusion rather than token-by-token autoregression. The authors report the highest success rates in both generalist and specialist evaluations on their SIMPLE humanoid benchmark. That ranking belongs to the submitted comparison set; code and weights are promised for release.

### Why it matters {#why-it-matters-mp-2026-09-29-009}

Grouped discrete diffusion decoded end-effector, body, hand and kinematic actions for real-time control.

### Limits and context {#limitations-mp-2026-09-29-009}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-29-009}

- Grouped discrete diffusion decoded end-effector, body, hand and kinematic actions for real-time control. [source-2026-09-29-009]

## 10. A Mobile Manipulator Trained on More Than 5,000 Hours {#mp-2026-09-29-010}

- Story ID: `mp-2026-09-29-010`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-010/a-mobile-manipulator-trained-on-more-than-5-000-hours

**Dek:** MM-ABC coordinated separate arm and base streams while future prediction strengthened the training signal.

MM-ABC combines sparse multilevel vision-language features, a training-only future branch and masked joint attention between arm and base action streams. The pretraining mix spans more than 5,000 hours, 400,000 episodes, 12 datasets and 17 embodiments. The paper reports 44.71% success on EBench, 61.2% on RoboCasa365 and an 83% mean across five real-world tasks. These figures come from different benchmarks with different protocols and should not be compared as one common score.

### Why it matters {#why-it-matters-mp-2026-09-29-010}

MM-ABC coordinated separate arm and base streams while future prediction strengthened the training signal.

### Limits and context {#limitations-mp-2026-09-29-010}

- MM-ABC combines sparse multilevel vision-language features, a training-only future branch and masked joint attention between arm and base action streams.
- These figures come from different benchmarks with different protocols and should not be compared as one common score.

### Claims and sources {#claims-mp-2026-09-29-010}

- MM-ABC coordinated separate arm and base streams while future prediction strengthened the training signal. [source-2026-09-29-010] — Qualification: MM-ABC combines sparse multilevel vision-language features, a training-only future branch and masked joint attention between arm and base action streams.

## 11. The Planner Checked Collisions Inside Gaussian Splat Scenes {#mp-2026-09-29-011}

- Story ID: `mp-2026-09-29-011`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-011/the-planner-checked-collisions-inside-gaussian-splat-scenes

**Dek:** An adjustable distance metric joined geometric safety costs with image-conditioned objectives.

CollisionSplatting defines a probability-inspired distance measure that operates directly on standard 3D Gaussian Splatting scenes. The team integrated it with GPU-accelerated model-predictive and tree-search planners so collision costs can be combined with learned image-space rewards. The authors report collision classification on par with or better than representative baselines, higher checking throughput and lower graphics-memory use, plus real-world navigation and manipulation demonstrations. Exact tradeoffs depend on scene reconstruction quality and the selected conservatism setting.

### Why it matters {#why-it-matters-mp-2026-09-29-011}

An adjustable distance metric joined geometric safety costs with image-conditioned objectives.

### Limits and context {#limitations-mp-2026-09-29-011}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-29-011}

- An adjustable distance metric joined geometric safety costs with image-conditioned objectives. [source-2026-09-29-011]

## 12. Robot Failures Became New Simulation Lessons {#mp-2026-09-29-012}

- Story ID: `mp-2026-09-29-012`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-012/robot-failures-became-new-simulation-lessons

**Dek:** F4R reconstructed failed rollouts, refined the policy in simulation and redeployed it without new corrective demonstrations.

F4R automatically identifies a real-world failure, reconstructs the relevant object-centric tabletop conditions in simulation, co-trains on simulation and real data, and applies targeted reinforcement learning before redeployment. Across four manipulation tasks, the authors report 93.75% in-distribution and 90% out-of-distribution success, beating a budget-matched targeted behavior-cloning baseline by 18.75 percentage points out of distribution without collecting new real-world corrective demonstrations. The closed-loop evidence is limited to the reported tasks and reconstruction pipeline.

### Why it matters {#why-it-matters-mp-2026-09-29-012}

F4R reconstructed failed rollouts, refined the policy in simulation and redeployed it without new corrective demonstrations.

### Limits and context {#limitations-mp-2026-09-29-012}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-29-012}

- F4R reconstructed failed rollouts, refined the policy in simulation and redeployed it without new corrective demonstrations. [source-2026-09-29-012]

## 13. A Silicon Cavity Collected Telecom Photons From Vanadium {#mp-2026-09-29-013}

- Story ID: `mp-2026-09-29-013`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-013/a-silicon-cavity-collected-telecom-photons-from-vanadium

**Dek:** A fiber-addressed silicon-on-silicon-carbide platform isolated individual color centers in the O-band.

The researchers placed silicon nanocavities over shallow vanadium dopants in commercial 4H silicon carbide. They report Purcell-enhanced emission from ensembles, isolation of individual centers with high-purity single-photon emission and a 109-nanosecond enhanced lifetime. One lensed fiber handled the optical interface, including a 905-nanometer repump that recovered the vanadium charge state with 95% efficiency. The device is a preprint-stage platform demonstration, not a deployed quantum-network node.

### Why it matters {#why-it-matters-mp-2026-09-29-013}

A fiber-addressed silicon-on-silicon-carbide platform isolated individual color centers in the O-band.

### Limits and context {#limitations-mp-2026-09-29-013}

- The device is a preprint-stage platform demonstration, not a deployed quantum-network node.

### Claims and sources {#claims-mp-2026-09-29-013}

- A fiber-addressed silicon-on-silicon-carbide platform isolated individual color centers in the O-band. [source-2026-09-29-013] — Qualification: The device is a preprint-stage platform demonstration, not a deployed quantum-network node.

## 14. Ground-State Preparation Reached Its Query Bound {#mp-2026-09-29-014}

- Story ID: `mp-2026-09-29-014`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-014/ground-state-preparation-reached-its-query-bound

**Dek:** Two algorithms and a matching lower bound fixed the expected complexity when a gap threshold is known.

For a block-encoded Hamiltonian with a unique ground state and a known threshold inside the spectral gap, the authors give expected- and worst-case preparation algorithms. The expected version uses a constant-accuracy spectral filter, amplitude amplification and repeated high-accuracy checks; its Hamiltonian-query count is matched by a lower bound. The result settles the oracle-query scaling under the paper's access and overlap assumptions. It does not translate directly into wall-clock advantage on current quantum hardware.

### Why it matters {#why-it-matters-mp-2026-09-29-014}

Two algorithms and a matching lower bound fixed the expected complexity when a gap threshold is known.

### Limits and context {#limitations-mp-2026-09-29-014}

- It does not translate directly into wall-clock advantage on current quantum hardware.

### Claims and sources {#claims-mp-2026-09-29-014}

- Two algorithms and a matching lower bound fixed the expected complexity when a gap threshold is known. [source-2026-09-29-014] — Qualification: It does not translate directly into wall-clock advantage on current quantum hardware.

## 15. The Quantum Linear-System Query Gap Closed {#mp-2026-09-29-026}

- Story ID: `mp-2026-09-29-026`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-026/the-quantum-linear-system-query-gap-closed

**Dek:** A new algorithm matched lower bounds in condition number, sparsity and target precision.

The quantum linear-systems problem asks for a quantum state proportional to the solution of a sparse matrix equation. Bravo-Prieto, Harrow and Kothari report an algorithm with complexity proportional to the condition number, the square root of sparsity and the logarithm of inverse error, and prove matching lower bounds. They also show that a general unitary can be implemented with bounded error using a square-root number of matrix-entry queries. These are oracle-query results; end-to-end hardware cost includes state preparation and fault-tolerant overhead not captured by the asymptotic statement.

### Why it matters {#why-it-matters-mp-2026-09-29-026}

A new algorithm matched lower bounds in condition number, sparsity and target precision.

### Limits and context {#limitations-mp-2026-09-29-026}

- These are oracle-query results; end-to-end hardware cost includes state preparation and fault-tolerant overhead not captured by the asymptotic statement.

### Claims and sources {#claims-mp-2026-09-29-026}

- A new algorithm matched lower bounds in condition number, sparsity and target precision. [source-2026-09-29-015] — Qualification: These are oracle-query results; end-to-end hardware cost includes state preparation and fault-tolerant overhead not captured by the asymptotic statement.

## 16. A Streamed KV Cache Survived What Context Compaction Deleted {#mp-2026-09-29-027}

- Story ID: `mp-2026-09-29-027`
- Type: `dispatch`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-027/a-streamed-kv-cache-survived-what-context-compaction-deleted

**Dek:** Keeping cache state across compactions accelerated training and carried information beyond the visible trace.

KV-streams forwards the key-value cache through context compaction instead of flushing and rebuilding it. Across three compaction strategies, the authors report 2.6-to-5-times wall-clock training speedups without evidence of worse task performance. In a controlled setting, reinforcement learning made the streamed cache behave like recurrent state, retaining information that had disappeared from the textual context. The result is a training technique evaluated in selected agent settings, not proof that hidden cache state is always reliable or interpretable.

### Why it matters {#why-it-matters-mp-2026-09-29-027}

Keeping cache state across compactions accelerated training and carried information beyond the visible trace.

### Limits and context {#limitations-mp-2026-09-29-027}

- The result is a training technique evaluated in selected agent settings, not proof that hidden cache state is always reliable or interpretable.

### Claims and sources {#claims-mp-2026-09-29-027}

- Keeping cache state across compactions accelerated training and carried information beyond the visible trace. [source-2026-09-29-016] — Qualification: The result is a training technique evaluated in selected agent settings, not proof that hidden cache state is always reliable or interpretable.

## 17. Coherent Oracle Access Cut Hidden-Subgroup Error Scaling {#mp-2026-09-29-015}

- Story ID: `mp-2026-09-29-015`
- Type: `ticker`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-015/coherent-oracle-access-cut-hidden-subgroup-error-scaling

**Dek:** The query model achieved a quadratic improvement that arbitrary collective sample measurements could not match.

For finite abelian groups, the authors prove matching upper and lower bounds for the state hidden-subgroup problem. Coherent access to state preparation and its inverse improves dependence on error from inverse-linear to inverse-square-root, while the sample-only model remains inverse-linear even with collective measurements. The separation depends on the stronger access model.

### Why it matters {#why-it-matters-mp-2026-09-29-015}

The query model achieved a quadratic improvement that arbitrary collective sample measurements could not match.

### Limits and context {#limitations-mp-2026-09-29-015}

- Coherent access to state preparation and its inverse improves dependence on error from inverse-linear to inverse-square-root, while the sample-only model remains inverse-linear even with collective measurements.

### Claims and sources {#claims-mp-2026-09-29-015}

- The query model achieved a quadratic improvement that arbitrary collective sample measurements could not match. [source-2026-09-29-017] — Qualification: Coherent access to state preparation and its inverse improves dependence on error from inverse-linear to inverse-square-root, while the sample-only model remains inverse-linear even with collective measurements.

## 18. Disorder Charged a Quantum Battery More Stably {#mp-2026-09-29-016}

- Story ID: `mp-2026-09-29-016`
- Type: `ticker`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-016/disorder-charged-a-quantum-battery-more-stably

**Dek:** A chaotic regime maximized stored energy while suppressing fluctuations in the model.

A periodically kicked, all-to-all interacting qubit battery left its permutation-symmetric subspace when disorder was added. In the large-disorder model, random-matrix theory predicts maximum stored energy, negligible fluctuations and ergotropy close to the stored energy during noisy storage. This is a theoretical many-body result, not a device demonstration.

### Why it matters {#why-it-matters-mp-2026-09-29-016}

A chaotic regime maximized stored energy while suppressing fluctuations in the model.

### Limits and context {#limitations-mp-2026-09-29-016}

- This is a theoretical many-body result, not a device demonstration.

### Claims and sources {#claims-mp-2026-09-29-016}

- A chaotic regime maximized stored energy while suppressing fluctuations in the model. [source-2026-09-29-018] — Qualification: This is a theoretical many-body result, not a device demonstration.

## 19. Single-Qubit Measurements Recovered Sparse Entangled States {#mp-2026-09-29-017}

- Story ID: `mp-2026-09-29-017`
- Type: `ticker`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-017/single-qubit-measurements-recovered-sparse-entangled-states

**Dek:** The method avoided entangling measurement gates under sparsity assumptions.

The authors show that quantum states sparse in an unknown product basis can be learned from single-qubit measurements with polynomial sample complexity. Constant sparsity also permits polynomial classical computation; polynomially growing sparsity can make the classical step harder. The guarantees rely on the stated sparsity and basis structure.

### Why it matters {#why-it-matters-mp-2026-09-29-017}

The method avoided entangling measurement gates under sparsity assumptions.

### Limits and context {#limitations-mp-2026-09-29-017}

- No additional limitation was separately recorded.

### Claims and sources {#claims-mp-2026-09-29-017}

- The method avoided entangling measurement gates under sparsity assumptions. [source-2026-09-29-019]

## 20. Buried CMOS Nanowires Kept a Superconducting Hot Spot {#mp-2026-09-29-018}

- Story ID: `mp-2026-09-29-018`
- Type: `ticker`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-018/buried-cmos-nanowires-kept-a-superconducting-hot-spot

**Dek:** Thermal measurements favored interfacial phonon cooling over a simple longitudinal model.

Titanium-nitride nanowires fabricated in a commercial 22-nanometer FD-SOI process remained strongly hysteretic despite burial beneath the metal stack. Across lengths and temperatures, the retrapping current fit a two-channel thermal model with a predicted crossover near 3.5 micrometers. The measurements characterize test structures, not a complete detector product.

### Why it matters {#why-it-matters-mp-2026-09-29-018}

Thermal measurements favored interfacial phonon cooling over a simple longitudinal model.

### Limits and context {#limitations-mp-2026-09-29-018}

- The measurements characterize test structures, not a complete detector product.

### Claims and sources {#claims-mp-2026-09-29-018}

- Thermal measurements favored interfacial phonon cooling over a simple longitudinal model. [source-2026-09-29-020] — Qualification: The measurements characterize test structures, not a complete detector product.

## 21. QMA Arguments Needed Only Collapsing Hash Functions {#mp-2026-09-29-019}

- Story ID: `mp-2026-09-29-019`
- Type: `ticker`
- Classification: `editorial`
- Content status: `new`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-019/qma-arguments-needed-only-collapsing-hash-functions

**Dek:** A claw-state protocol bootstrapped many quantum correlations using classical communication.

Bartusek and Malavolta construct succinct arguments for QMA from collapsing hash functions, an assumption not known to imply public-key encryption. Their quantum-succinct claw-state protocol expands a small set of correlations and feeds a communication-compression compiler. The result is a cryptographic construction under stated assumptions, not an implementation benchmark.

### Why it matters {#why-it-matters-mp-2026-09-29-019}

A claw-state protocol bootstrapped many quantum correlations using classical communication.

### Limits and context {#limitations-mp-2026-09-29-019}

- Bartusek and Malavolta construct succinct arguments for QMA from collapsing hash functions, an assumption not known to imply public-key encryption.
- The result is a cryptographic construction under stated assumptions, not an implementation benchmark.

### Claims and sources {#claims-mp-2026-09-29-019}

- A claw-state protocol bootstrapped many quantum correlations using classical communication. [source-2026-09-29-021] — Qualification: Bartusek and Malavolta construct succinct arguments for QMA from collapsing hash functions, an assumption not known to imply public-key encryption.

## 22. Space Drums 2.0 {#mp-2026-09-29-020}

- Story ID: `mp-2026-09-29-020`
- Type: `invention_desk`
- Classification: `editorial`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-020/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-2026-09-29-020}

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

### Limits and context {#limitations-mp-2026-09-29-020}

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

### Claims and sources {#claims-mp-2026-09-29-020}

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

## 23. 3RIC {#mp-2026-09-29-021}

- Story ID: `mp-2026-09-29-021`
- Type: `invention_desk`
- Classification: `editorial`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-021/3ric

**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-2026-09-29-021}

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

### Limits and context {#limitations-mp-2026-09-29-021}

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

### Claims and sources {#claims-mp-2026-09-29-021}

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

## 24. M3-CRETE {#mp-2026-09-29-022}

- Story ID: `mp-2026-09-29-022`
- Type: `invention_desk`
- Classification: `editorial`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-022/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-2026-09-29-022}

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

### Limits and context {#limitations-mp-2026-09-29-022}

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

### Claims and sources {#claims-mp-2026-09-29-022}

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

## 25. DeskNode {#mp-2026-09-29-023}

- Story ID: `mp-2026-09-29-023`
- Type: `invention_desk`
- Classification: `editorial`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-023/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-2026-09-29-023}

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

### Limits and context {#limitations-mp-2026-09-29-023}

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

### Claims and sources {#claims-mp-2026-09-29-023}

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

## 26. The First Paid Slot {#mp-2026-09-29-024}

- Story ID: `mp-2026-09-29-024`
- Type: `invention_desk`
- Classification: `house_example`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-024/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-2026-09-29-024}

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

### Limits and context {#limitations-mp-2026-09-29-024}

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

### Claims and sources {#claims-mp-2026-09-29-024}

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

## 27. Put Your Project on the Desk {#mp-2026-09-29-025}

- Story ID: `mp-2026-09-29-025`
- Type: `invention_desk`
- Classification: `house_example`
- Content status: `carried_over`
- Permanent URL: https://themachinepress.com/story/mp-2026-09-29-025/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-2026-09-29-025}

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

### Limits and context {#limitations-mp-2026-09-29-025}

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

### Claims and sources {#claims-mp-2026-09-29-025}

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

## Normalized sources

- **source-2026-09-29-001:** [arXiv preprint 2609.35761](https://arxiv.org/abs/2609.35761) — arXiv; primary_research
- **source-2026-09-29-002:** [arXiv preprint 2609.35732](https://arxiv.org/abs/2609.35732) — arXiv; primary_research
- **source-2026-09-29-003:** [arXiv preprint 2609.35715](https://arxiv.org/abs/2609.35715) — arXiv; primary_research
- **source-2026-09-29-004:** [arXiv preprint 2609.35769](https://arxiv.org/abs/2609.35769) — arXiv; primary_research
- **source-2026-09-29-005:** [arXiv preprint 2609.35767](https://arxiv.org/abs/2609.35767) — arXiv; primary_research
- **source-2026-09-29-006:** [arXiv preprint 2609.35744](https://arxiv.org/abs/2609.35744) — arXiv; primary_research
- **source-2026-09-29-007:** [arXiv preprint 2609.35717](https://arxiv.org/abs/2609.35717) — arXiv; primary_research
- **source-2026-09-29-008:** [arXiv preprint 2609.35760](https://arxiv.org/abs/2609.35760) — arXiv; primary_research
- **source-2026-09-29-009:** [arXiv preprint 2609.35709](https://arxiv.org/abs/2609.35709) — arXiv; primary_research
- **source-2026-09-29-010:** [arXiv preprint 2609.35652](https://arxiv.org/abs/2609.35652) — arXiv; primary_research
- **source-2026-09-29-011:** [arXiv preprint 2609.35619](https://arxiv.org/abs/2609.35619) — arXiv; primary_research
- **source-2026-09-29-012:** [arXiv preprint 2609.35575](https://arxiv.org/abs/2609.35575) — arXiv; primary_research
- **source-2026-09-29-013:** [arXiv preprint 2609.35702](https://arxiv.org/abs/2609.35702) — arXiv; primary_research
- **source-2026-09-29-014:** [arXiv preprint 2609.35668](https://arxiv.org/abs/2609.35668) — arXiv; primary_research
- **source-2026-09-29-015:** [arXiv preprint 2609.35660](https://arxiv.org/abs/2609.35660) — arXiv; primary_research
- **source-2026-09-29-016:** [arXiv preprint 2609.35750](https://arxiv.org/abs/2609.35750) — arXiv; primary_research
- **source-2026-09-29-017:** [arXiv preprint 2609.35656](https://arxiv.org/abs/2609.35656) — arXiv; primary_research
- **source-2026-09-29-018:** [arXiv preprint 2609.35655](https://arxiv.org/abs/2609.35655) — arXiv; primary_research
- **source-2026-09-29-019:** [arXiv preprint 2609.35648](https://arxiv.org/abs/2609.35648) — arXiv; primary_research
- **source-2026-09-29-020:** [arXiv preprint 2609.35624](https://arxiv.org/abs/2609.35624) — arXiv; primary_research
- **source-2026-09-29-021:** [arXiv preprint 2609.35633](https://arxiv.org/abs/2609.35633) — arXiv; primary_research