30
SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research?
SAEScientist-Bench tests whether AI agents can autonomously run SAE interpretability research on Gemma-2-9B-IT; frontier agents trail expert baselines.
The benchmark requires agents to design contrastive probes and navigate the Gemma Scope dictionary of over 131K features in Gemma-2-9B-IT to discover optimal interpretable features, scored against expert-curated references on Neuronpedia via activation rank, concept selectivity, and causal steering. Across 10 agent configurations and 20 tasks, frontier agents demonstrate genuine discovery capability and approach expert levels at separating target concepts from controls, but lag substantially in causal steering and frequently misinterpret experimental measurements. The authors frame this as establishing experimental model understanding as a measurable capability for closed-loop autonomous AI R&D and post-hoc monitoring for recursive self-improvement.
40
SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research?
SAEScientist-Bench evaluates whether AI agents can autonomously conduct SAE interpretability research in Gemma-2-9B-IT, finding frontier agents trail expert baselines.
SAEScientist-Bench tests if AI agents can act as scientists using SAE tools for autonomous mechanistic discovery, requiring them to design contrastive probes and navigate a Gemma Scope dictionary of 131K+ features in Gemma-2-9B-IT. Across 10 agent configurations and 20 tasks, frontier agents showed genuine discovery capability but remained well behind expert reference features, lagging most in causal steering. Agents frequently misinterpreted experimental measurements even when designing effective contrasts.
55
Det-LIME: Detector-Aware, Multi-Instance Local Interpretable Model-Agnostic Explanations for Automated Marine Mammal Detection
Det-LIME extends LIME to multi-instance object detection explanations, improving attribution for harbor seal aerial surveys.
Det-LIME adapts LIME to object detection by combining per-detection weighting, a proximity kernel emphasizing box-adjacent regions, and IoU-based matching to track instances across perturbations. It was evaluated on aerial drone imagery for harbor seal detection plus a seabird case study, and compared against vanilla LIME, Stabilized LIME, Deterministic LIME, and gradient-based attribution. Using Attribution Ratio and Max Saliency Hit Rate metrics, it consistently improved multi-instance attribution and produced box-aligned explanations useful for debugging and data augmentation.
15
Explainability Assistant: A Conversational XAI Interface for Interpreting Energy Consumption Models
Researchers release Explainability Assistant, an open-source conversational XAI tool using LLM function calling, lifting intent-parsing accuracy from 76.8% to 94%.
The paper introduces the Explainability Assistant, an open-source conversational XAI system for interpreting energy consumption forecasting models such as genetic-programming symbolic regressors. It uses LLM function calling instead of rigid custom grammars, achieving 94% intent-parsing accuracy versus 76.8% for prior work TalkToModel, and adapts to different ML problem types without task-specific fine-tuning. Comparative evaluation with energy domain specialists against a traditional XAI dashboard showed improved usability, with all experts preferring the conversational interface.
25
AlphaGenome Atlas: A predictive map of every possible DNA letter change in the human genome
Google DeepMind released AlphaGenome Atlas, a free 1-petabyte platform predicting the molecular effects of all ~9 billion possible single-letter DNA variants.
Google DeepMind introduced AlphaGenome Atlas, containing precomputed predictions for the effects of roughly 9 billion single-nucleotide variants across the human genome, spanning hundreds of human and mouse cell types. The 1-petabyte dataset is more than 30 times larger than the AlphaFold Database and includes an AlphaGenome Variant Impact (AVI) score combining AlphaGenome and AlphaMissense predictions for both coding and non-coding regions. External collaborators have already used it to identify and experimentally verify variants in unsolved rare disease research. It is available via a free web portal, the AlphaGenome API, and as a skill in Google Antigravity.
42
CARDEA: Auditable Reasoning Grounded in Spatial Evidence for End-to-End Coronary Angiography Interpretation
CARDEA, a vision-language model trained only on public data, matches cardiologists on coronary angiography complexity assessment while exposing auditable bounding-box evidence.
CARDEA is a unified large vision-language model serving as the inference core of an end-to-end coronary angiography pipeline from multi-view videos to study-level diagnosis. It was trained on public datasets through visual alignment, self-distilled Chain-of-Box cold start, and reinforcement learning with verifiable rewards encouraging bounding-box reasoning. It reached 0.91 accuracy on dominance classification under domain shift and 0.90 on complexity assessment, comparable to two interventional cardiologists. RLVR raised zero-shot report generation vessel-severity macro-F1 from 0.513 to 0.686, while supervised imitation alone did not.
30
Beneath the Surface of Chains-of-Thought: A Mechanistic Interpretation of Reasoning Operations in LLMs
Study shows LLM reasoning operations like planning and deduction are geometrically separable in hidden states, with separability peaking in middle layers.
Researchers investigate whether functional reasoning operations — problem formulation, goal decomposition, deduction — have corresponding geometric structure in LLM hidden representations. They find operations are separable in held-out representations with separability peaking in middle layers, ruling out lexical and positional confounds; token-wise operation alignment becomes more distributed across layers, and identical surface tokens are represented differently depending on their surrounding chunk. Attention-masking interventions show chunk-onset operation-aligned representations depend on preceding reasoning context; code is released on GitHub (naver-ai/beneath-cot).
40
30
Local gradient neural operator
Researchers propose LGNO, a lightweight interpretable neural operator using learnable local stencils, matching global-operator accuracy on PDE benchmarks with fewer parameters.
LGNO builds on nonlinear gradient discretization priors and uses multilayer perceptron convolutional layers to learn translation-invariant local kernels resembling discrete stencils. A zero consistent stencil factorization separates coefficient learning from field reconstruction, and network folding shares equivalent components to cut parameter counts for symmetric problems. Evaluations on linear and nonlinear, static and dynamic, and low- and high-dimensional PDE benchmarks show maintained accuracy, parameter efficiency, and rollout stability, with applicability to diffusion, flow, and quantum problems.
28
[AINews] OpenAI reports Navier-Stokes singularity find in 88 hours using Astra-next, roughly 10,000 agents and 130B tokens (>$40M), a contender for second ever Millennium Prize awarded
OpenAI-linked accounts claim roughly 10,000 AI agents produced a Navier-Stokes singularity result in 88 hours, pending mathematical verification.
OpenAI-affiliated accounts claim a system of roughly 10,000 agents, trained over about a year with multi-agent reinforcement learning, produced a finite-time singularity result related to the Navier-Stokes Millennium Problem. The claimed 88-hour runtime and 130B-token cost circulate only via social posts, and no preprint, theorem statement, or proof artifact is available. Acceptance by the mathematics community is unresolved, so the claim's epistemic status remains unknown. The roundup also notes Cognition's $48B and Mistral's $24B fundraises, GPT Image 2.5, and Meta's Muse agent relaunch.
55
When LLM judges agree, should we believe them?
Amazon ICML paper uses Ising models to correct correlated LLM-judge votes, beating accuracy-weighted panels by 9-14%.
Amazon Science describes an ICML paper, "Dependence-aware label aggregation for LLM-as-a-judge via Ising models," addressing how correlated judge outputs inflate majority-vote confidence. The unsupervised method models pairwise dependence between judges, learning both reliability and similarity without human reference labels. Tested on relevance, toxicity, and summarization tasks with 10 judge models at temperature zero, it outperformed accuracy-weighted voting by 9% to 14%.
Hacker News · AI40
Omni-Streaming Thinking
Omni-Streaming Thinking fixes premature cross-modal commitment in streaming omni-modal models via pending claims verified against modality-specific evidence, beating baselines by over 10%.
The paper identifies 'premature cross-modal commitment', where streaming models keep relaying early visual interpretations even after audio contradicts them. OST generates evidence-linked pending claims with future verification intervals, stores audio and visual evidence separately, and refutes claims when contradictory evidence appears. Built on a frozen Qwen3-Omni-30B-A3B-Instruct backbone with lightweight adaptation, it outperforms open baselines by more than 10% relative on five streaming and audio-visual benchmarks. On the new OST-DiagBench it reaches d-prime 2.95 versus at most 1.38 for open baselines, while reducing vision-induced auditory hallucinations.
45
MAxBench: A Multinomial Concept Recovery Benchmark
MAxBench evaluates multinomial concept recovery methods, finding affine subspaces steer most reliably but none consistently beats prompting.
MAxBench is a geometry-agnostic evaluation framework for multinomial concept representations in language models, based on sampling from recovered concept representations. It compares 10 localization methods covering 5 geometry types across 6 concepts and 4 models. Findings show affine subspaces steer more reliably than rank-one or linear subspaces due to better non-zero offsets, manifold steering is competitive where applicable, and no method consistently outperforms prompting.
18
From Parameters to Answers: How LLMs Retrieve and Use Their Internal Knowledge
Interpretability study traces how Qwen, Llama, and Gemma route query information and internal knowledge across layers when answering questions.
Researchers used layerwise interventions on hidden states to separate query-routing signals from target knowledge in language models, testing Qwen, Llama, and Gemma on country-continent questions with varied answer types. A pair-conditioned request direction strengthens before interventions alter downstream knowledge, opening a causal window while answer-supporting content is still forming. Trajectories differ by model: Gemma shows a partially overlapping mid-layer routing profile, while Llama has no sustained routing-effect window under the same gates.
25
Google DeepMind Releases AlphaGenome Atlas With Precomputed Molecular Effect Predictions and AVI Scores for 9 Billion Human DNA Variants
Google DeepMind launched AlphaGenome Atlas, precomputing molecular effect predictions and AVI impact scores for ~9 billion human single-nucleotide variants in a 1-petabyte catalogue.
Google DeepMind released AlphaGenome Atlas, a 1-petabyte catalogue of precomputed molecular effect predictions for roughly 9 billion possible single-nucleotide variants in the human genome. It introduces the AlphaGenome Variant Impact (AVI) score, combining AlphaGenome regulatory predictions with AlphaMissense, plus per-variant feature attributions and over 2,500 recurrent DNA sequence motifs. DeepMind reports best-in-class AVI performance on variant pathogenicity and rare disease benchmarks. Early users at the Broad Institute, University of Exeter, and Stowers Institute demonstrated rare-disease variant reprioritization and 22% more non-coding associations across 54,000+ UK Biobank genomes.
55
VoT: Vision-of-Thought for Unified Multimodal Representation Alignment
Researchers propose Vision-of-Thought (VoT), a discrete visual-planning token layer between VLMs and diffusion transformers improving text-to-image semantic alignment.
VoT introduces a discrete visual-thinking layer between vision-language models and diffusion transformers, letting the VLM act as a multimodal planner that emits tokens describing objects and layouts before pixel generation. A specialized VoT tokenizer is trained with VLM alignment, feature reconstruction, and vector-quantization losses. Experiments show improved semantic alignment and a structured, interpretable interface for controllable generation.
24
Who Should Grade My Work? Student Perspectives on Transparent AI-Assisted Writing Assessment in Higher Education
A Saudi university study finds students value ChatGPT writing feedback but treat human instructors as the final grading authority.
Thirteen male undergraduate computing students at a Saudi public university completed handwritten writing tasks that were scored by ChatGPT using a rubric-based prompt, then reflected after being told the score and feedback were AI-generated. Inductive thematic analysis identified four themes: perceived feedback usefulness, awareness of AI's contextual and pedagogical limitations, conditional trust, and reflection on the instructor's institutional role. Participants accepted GenAI feedback for surface-level revision but consistently positioned human instructors as the authority over grading decisions, distinguishing feedback utility from evaluative authority.
15
Technical Manual for a Toolkit for Measuring Contextual Individuation in Transformer Language Models
An open methodology toolkit measures whether transformer language models contextualize fixed word forms across domains using bridge forms and layer-wise silhouette analysis.
The manual documents an open toolkit built around 'bridge forms' - identical written words recurring across two or more subject domains with a different sense in each - to test whether transformer language models individuate word occurrences by context beyond the embedding layer. It covers declarative specification of bridge forms, Wikipedia corpus acquisition, occurrence localization, layer-wise representation extraction, domain-pairwise silhouette measurement, and visualization, justifying each choice against failure modes such as sense contamination and subword-tokenization misalignment. It is a methodological and implementation reference and reports no empirical results.
14
Unfold The World: Factorize 4D Properties in Reinforcing Spatial Reasoning
FactoSR factorizes 4D spatial reasoning into XY, Z, and T reinforcement-learning sub-objectives, boosting VLM performance on VSI-Bench by 5.9% and All-Angles-Bench by 4.5%.
Researchers present FactoSR, a factorized reinforcement learning framework that decomposes world-consistent reasoning into planar correspondence, depth consistency, and temporal reversibility sub-objectives. Optimizing these verifiable constraints turns the ill-posed projection recovery problem into tangible reasoning steps. Evaluations show gains of 5.9% on VSI-Bench and 4.5% on All-Angles-Bench for 3D and 4D reasoning, arguing VLMs' spatial bottleneck stems from training on 2D projections versus latent 3D geometry and temporal continuity.
26
MasterControl Seventeen Every Time
Governed enterprise analytics study shows deterministic policy execution matched 110/110 answer-and-evidence contracts while runtime agent planning matched none.
The paper studies a governed approach where a language model interprets the question while deterministic policy selects and runs a pre-approved analytical program returning results and evidence. Across 440 runs, three 8B models generated SQL and selected tools at runtime, while Qwen3-8B only interpreted intent and policy executed the approved program. None of 330 runtime-planning episodes satisfied the full answer-and-evidence contract, whereas the policy-executed analyzer matched 110 of 110. The authors note this is configuration-specific and expressiveness is preserved via relational operations, aggregation, comparison, windows, ranking, and similarity with replayable results.
24
30
30
30
Coding Agents Have Converged: Why the SWE-bench Leaderboard Can No Longer Order Its Top Entries, and What to Measure Instead
Audit of 254 SWE-bench submissions finds top coding-agent entries statistically inseparable, so small leaderboard gaps no longer establish rank.
The paper audits 254 SWE-bench submissions across four splits without running models. On Verified, the top two entries each resolve 396 of 500 instances, and exact paired McNemar tests separate none of the 29 adjacent top-thirty pairs at alpha=0.05. Within-model scaffold score ranges reach 29.8 percentage points, versus an 8.8-point spread among the top thirty. The authors release a five-step audit protocol and recommend reporting comparison-set-specific resolution and model-scaffold provenance.
40
Task-Directed Residual AddUNet:Perfect-Reconstruction Routing for Full-Rate Representations
Paper recasts additive U-Net skip structure as a perfect-reconstruction filter bank and adds full-rate residual routing for task-directed representations.
The work proves a constrained additive U-Net's survivor-skip structure is exactly equivalent to a critically sampled perfect-reconstruction filter bank, and removes complementary-subband restrictions via a full-rate formulation. A Residual Full-Rate PR architecture routes task-irrelevant or redundant structure away from the task pathway while guaranteeing exact reconstruction without invertible operators, a matched synthesis bank, or a learned decoder. On TIMIT, the front-end improves test PER from 28.60±2.09% to 25.76±0.41% with recognizer and training held fixed.
12
HarnessVLN: Unifying Training-Free Embodied Navigation through an Agent Harness
Researchers introduce HarnessVLN, a zero-shot training-free agent harness that sets new training-free SOTA on vision-language navigation benchmarks including R2R and HM3D.
HarnessVLN is a zero-shot, training-free framework for embodied vision-language navigation that coordinates perception, retrieval, grounding, navigation, recovery, and termination through a unified tool interface. It validates planner proposals against spatial evidence, geometric feasibility, and subgoal consistency, using hierarchical event memory and a persistent Spatiotemporal Graph that stores reusable spatial evidence and failure annotations. It reports success rates of 60.8% on R2R, 53.9% on RxR, 76.0% on HM3D-v2, and 59.3% on HM3D-OVON, surpassing prior training-free state of the art, with real-world humanoid deployment demonstrated.
30
Retrospectively Reverse-Engineering Apple's Neural Engine
A developer reverse-engineers Apple's M1 Neural Engine architecture, mapping compute cores, MAC datapaths, and schedulers to explain the NPU's decline as transformers displaced CNN workloads.
A developer who previously maintained a reverse-engineered Linux driver for Apple's Neural Engine (ANE) published a retrospective deep dive mapping the M1 ANE's full internal architecture: compute, datapath, scheduler, memory, and execution model. The M1 ANE has 16 compute cores with 128 FP16 (or 256 INT8) MAC lanes each, totaling 2048 parallel MAC lanes, using 32-bit Q16.16 fixed-point accumulation with FP16 readout and an accumulator that saturates at 2^15. The author argues the ANE's dataflow was architected around the predictable reuse patterns of 2017-era CNN workloads (dating to the A11 Bionic), which autoregressive transformer decode broke, limiting its usefulness for general ML. With Apple's M5 folding ANE cores into GPU cores to tout LLM performance, the post frames this as the beginning of the end for the standalone NPU.
28
StepAudio 3 Realtime Technical Report
StepAudio 3 Realtime debuts an audio-language model with Think-While-Speaking reasoning, delivering full-duplex voice dialogue with top benchmark results.
StepAudio 3 Realtime is an audio-language foundation model built around a continuous listen-converse-think-act loop for real-time spoken interaction. Think-While-Speaking runs private reasoning in parallel with speech, reaching a 73.0 macro average on StepAudioChat in reasoning mode. The model reports 90.6 on MMSU, 98.9 overall on the Artificial Analysis Full-Duplex Bench, and 56.0% macro task success on tau-Voice. An integrated Voice Agent handles asynchronous tool execution without disrupting dialogue flow.
45
Agent as Policy for Robotic Manipulation
Agent as Policy lets a general-purpose agent drive a physical robot via runtime reasoning and program generation, reaching 100% success on manipulation tasks.
The paper introduces Agent as Policy (AGP), which puts task planning and execution for a physical robot under a general-purpose agent's control with no task-specific or environment-specific training. The agent interprets visual evidence, writes executable programs, issues motion commands, and revises actions based on physical outcomes. AGP was evaluated on real-world manipulation tasks including assembly from human videos, block construction from goal images, die reorientation, targeted throwing, and bimanual towel folding. It achieved success rates of 100%, 100%, and 80% on three block construction configurations.
40
CausalArena: Benchmarking Causal Discovery in the Foundation Model Era
Researchers introduce CausalArena, a unified benchmark revealing that causal discovery rankings shift substantially across structural causal model families and protocols.
The paper presents CausalArena, a unified and evolvable benchmark for causal discovery combining synthetic structural causal models, semantically grounded operational SCMs, formula-grounded scientific SCMs, and public real-world datasets. Experiments across classical, neural, and pretrained causal discovery foundation models show large ranking shifts between benchmark regimes. The authors identify pretraining-evaluation overlap and benchmark diversity as central evaluation challenges.
30
DRG-MAPPO: Hierarchical Dynamic Role-Graph Multi-Agent Reinforcement Learning for Cooperative Air Combat
DRG-MAPPO combines graph-based relational modeling with dynamic role assignment in multi-agent RL, reaching an 87% win rate in cooperative air combat.
The hierarchical framework uses graph attention to extract relational features among allies, enemies, and threats, with a high-level policy assigning tactical roles like leader and supporter. A low-level policy executes discrete maneuver actions conditioned on roles and graph features, plus a target-priority auxiliary task encouraging focus-fire behavior. Experiments report a state-of-the-art 87% win rate, balancing relational modeling, interpretability, and optimization stability.
20
Show-Harness: Just a VLM Agent Can Play Robots
Show-Harness lets VLM agents control robots via discrete semantic action units, outperforming VLA baselines zero-shot and after light fine-tuning.
Show-Harness is an embodied agent harness that exposes discrete semantic action units a VLM reasons over, with embodiment-specific interpreters grounding them into local robot actions. It enables zero-shot robot control with closed-source frontier VLMs and low-cost adaptation of small open-source VLMs using only a few GPU-hours of fine-tuning. The companion GUMI (GUI Manipulation Interface) extends the same semantic action space to GUI-based demonstration collection without specialized teleoperation hardware. Experiments show robust generalization across tasks, embodiments, and environments, beating representative agentic and VLA paradigms.
32
Cross-Model Agreement as a Deployment-Time Reliability Signal for Automatic Polyp Segmentation
Referee-Based Quality Estimation flags unreliable polyp segmentations at inference without ground truth, reaching ROC-AUC 0.960 with SegFormer-B0 referees.
RBQE measures agreement between a primary segmentation model and an independently trained referee on a 1,223-image external benchmark drawn from four public datasets. A cross-architecture SegFormer-B0 referee achieves the strongest signal (ROC-AUC 0.960), beating a Test-Time Augmentation baseline by 0.055 ROC-AUC under an identical protocol. Excluding trivially separable empty-mask cases, ROC-AUC falls to 0.876 (SegFormer-B0) and 0.783 (same-architecture control), but RBQE's margin over baselines widens. Progressive rejection of low-agreement predictions increases mean Dice of retained outputs, supporting selective prediction at the cost of one extra forward pass.
30
Algorithmic stability via ensembling
Theoretical work derives a general framework quantifying stability guarantees for averaging-based ensembles under arbitrary data perturbations via covariance operator norms.
The paper develops a framework for quantifying algorithmic stability of ensembling strategies defined via averaging, for varied types of data perturbation. The main result bounds the stability of the ensembled algorithm in terms of the norm of a covariance operator describing the ensembling process. The framework yields interpretable insights across practical perturbation examples and provides sharper guarantees than those derived from differential privacy considerations.
22
Show-Harness: Just a VLM Agent Can Play Robots
Show-Harness enables VLM agents to control robots via a semantic action interface, achieving zero-shot frontier control and few-GPU-hour adaptation of small VLMs.
Show-Harness exposes discrete semantic action units that VLMs reason over, with embodiment-specific interpreters deterministically grounding them into local robot actions. It enables zero-shot closed-source frontier VLM control and adapts small open-source VLMs for low-cost deployment with a few GPU-hours of fine-tuning. The companion GUMI interface extends the same semantic action space to GUI-based demonstration collection without teleoperation hardware, and Show-Harness-equipped agents outperform representative agentic and VLA paradigms.
28
SyncWorld: Visual Calibration Enables World Models as Zero-Shot Simulators
SyncWorld is an action-conditioned world model acting as a zero-shot robotics simulator across unseen environments via visual calibration.
Researchers propose SyncWorld, an action-conditioned world model that simulates robot action outcomes in unseen environments without additional training. It uses a visual calibration episode of paired frames and actions to establish the setup-specific Action-Visual Mapping in context. Experiments show accurate simulation of action outcomes in novel settings and that simulated rollouts enable test-time policy improvement without training.
25
What Does an LLM-Agent Leaderboard Rank Actually Compare?
A methodological study shows close LLM-agent leaderboard rank gaps on SWE-bench and similar benchmarks often do not support superiority claims.
The paper defines an estimand-aware pairwise procedure for comparing agents, checking common support and applying explicit uncertainty rules and practical margins. Across SWE-bench, AgentRewardBench, and tau2-bench, close rank differences are frequently unresolved, and proxy labels or utility rules can change which system is selected. The authors argue a leaderboard score summarizes a released evaluation but does not by itself justify pairwise superiority conclusions.
20
Scalability Analysis of Distributed Kolmogorov-Arnold Network Training on High-Performance Computing Systems
An empirical study shows distributed Kolmogorov-Arnold Network training reaches 74.7% parallel efficiency at 8 A100 GPUs, with overheads driven by All-Reduce choices.
The study evaluates data-parallel Kolmogorov-Arnold Network (KAN) training on the FinisTerrae III supercomputer using up to 8 NVIDIA A100 GPUs across 4 nodes with PyTorch Distributed Data Parallel. Strong scaling yields 5.97x speedup and 74.7% parallel efficiency at 8 GPUs, comparable to conventional deep learning workloads, while communication overhead ranges from 1.3% to 6.1%, driven mainly by All-Reduce algorithm selection and inter-node latency rather than KAN's edge-wise gradient structure. Weak scaling shows an initial single-to-multi-GPU throughput drop followed by stability, and the parameter-to-memory ratio improves with model size even as training time scales unfavorably. The authors provide GPU topology and model-size deployment guidelines for KAN training.
26
WearableQA: A Benchmark for Health Reasoning over Real-World Wearable Data
WearableQA benchmark tests LLM health reasoning over longitudinal wearable data; the best of 14 evaluated LLMs reaches 72.9% accuracy.
WearableQA comprises 4,084 ten-option multiple-choice questions built from wearable time series, blood biomarkers, and demographics of 200 real users with up to 500 days of daily measurements. It defines 16 question types along two axes: data versus health reasoning, and single- versus cross-signal reasoning. Evaluation of 14 proprietary and open-source LLMs shows performance from 19.6% to 72.9% against a 10% chance baseline, with most models below 60%.
14