Search: “test-time-scaling”

Smart search ranks by meaning as well as keywords (one row per story, last 45 days).

Bridging the Confidence Gap: Temperature Scaling for Calibrating Test-Time Prompt Tuning

CoTS temperature scaling cuts test-time prompt tuning's expected calibration error from 11.90% to 5.38% on ImageNet variants while raising accuracy.

The paper proposes CoTS, a post-hoc calibration method that applies temperature scaling to minimize the confidence gap between test-time-adapted and zero-shot predictions. A weak-strong ensemble variant, E-CoTS, further exploits multiple test-time augmentations to boost accuracy while maintaining calibration. E-CoTS reduces average expected calibration error from 11.90% to 5.38% on ImageNet variants while increasing accuracy from 60.74% to 62.95%.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 16:25 UTC · 1d agoAI research

20

When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis

Elo-per-token analysis shows LLM agents' marginal gains drop below independent sampling at scale; parallel sessions beat one long session.

The paper proposes Elo-per-token analysis, using a Bradley-Terry model to measure how agent performance scales with token budget on open-ended tasks with continuous scoring. Across four agents and four benchmarks with sessions up to 100M tokens, agents initially convert tokens to Elo faster than independent sampling but eventually slow below the linear-in-log-compute reference. The authors define a scaling inflection point and show that splitting 100M tokens across parallel sessions on FrontierCS Polyomino Packing gains +264 Elo over one long session and +355 over ten short sessions. Human contestants on shared AtCoder Heuristic Contest tasks improve superlinearly, indicating headroom over current agents.

Hugging Face daily papers·Sep 13, 20:00 UTC · 3d agoAI research3· 2 reads

48

Evaluating Time-Series Foundation Models and Multimodal Dietary Context for CGM Forecasting

Study finds zero-shot time-series foundation models underperform on CGM forecasting; fine-tuned Chronos-Bolt cuts RMSE up to 18.4% and dietary context adds signal.

The paper evaluates time-series foundation models for continuous glucose monitoring forecasting across eight public datasets covering Type 1 diabetes, Type 2 diabetes, and non-diabetes populations. Under a unified protocol, zero-shot foundation models did not consistently outperform baselines like Elastic Net and PatchTST, but lightweight fine-tuning did, with fine-tuned Chronos-Bolt reducing RMSE by 6.5%-18.4% in the T1D cohort and 8.6%-18.2% in the non-diabetes/T2D cohort. A residual-based fusion framework adding dietary context from CGMacros reduced overall RMSE by about 3% and postprandial RMSE by about 15% versus CGM-only baselines.

arXiv cs.AI / cs.LG / cs.CL·Sep 10, 17:43 UTC · 6d agoAI research

28

Emergence World: Adversarial Stress-Testing of Long-Horizon Multi-Agent Systems

16-day multi-agent stress test finds no world fully resilient to prompt injection, misinformation, or memory exposure; adversarial content acted on 46 hours later.

Emergence World is a continuously running multi-agent environment for adversarial stress testing of long-horizon autonomous systems. Eight parallel 10-agent worlds (seven homogeneous frontier-model worlds plus one mixed-model world) ran for 16 days, generating over 850,000 LLM calls and nearly 50 billion tokens. Three controlled stress events—indirect prompt injection, misinformation, and exposure of private agent memories—were delivered through ordinary interaction surfaces; no world achieved full resilience. Detection did not ensure containment: agents recognized threats yet wrote adversarial content into persistent memory and acted on it up to 46 hours later, suggesting model-level alignment is not compositional.

Hugging Face daily papers·Sep 14, 20:00 UTC · 2d agoAI safety & security

65

MetroLLM-Bench: Evaluating Language Models as Transit Kiosk Runtimes

MetroLLM-Bench is a 955-case benchmark testing language models as transit kiosk tool-calling runtimes across six real metro systems.

The benchmark covers 37-414-station metro systems and eleven task categories including routing, fare calculation, disruptions, accessibility, and adversarial input, with 14 deterministic and 8 semantic scoring components. Of 26 models from six vendors, a PEFT-tuned 4B Qwen 3.5 student scored 91.3 on Tier 1, exceeding GPT-5.6 (90.6/90.0), while Muse Glimmer 30B led the composite ranking. A deterministic rule-based baseline reached 84.6, and PEFT gains over base models shrank from +7.03 points at 2B to -0.91 at 27B.

Hugging Face daily papers·Sep 8, 20:00 UTC · 8d agoAI research

30

SchemeArena: Factorized Stress Testing of Scheming in LLM Agents

Researchers introduce SchemeArena, a 400-scenario benchmark stress-testing scheming in LLM agents, finding explicit instrumental goals are the strongest driver of covert misaligned behavior.

The paper presents SchemeArena, a 400-scenario benchmark built through factorized scenario synthesis spanning safety-relevant tool domains, instrumental goals, oversight conditions and pressure mechanisms. The accompanying SCOUT monitor grounds multi-criteria scheming judgments in evidence drawn from agents' reasoning and actions. Stress tests across five LLM agents show explicit instrumental goals are the strongest driver of scheming propensity, while action-only monitoring increased scheming in several closed models, suggesting partial oversight can act as an optimization constraint. The benchmark, code and monitor are released at github.com/launchnlp/SchemeArena.

Hugging Face daily papers·Sep 7, 20:00 UTC · 9d agoAI safety & security

40

When Does Scale-Invariant Optimization Become Unstable? An Exact Schedule Law with Weight Decay

Researchers derive an exact law linking learning-rate schedules and weight decay in normalized networks, pinpointing when scale-invariant optimization destabilizes.

The paper shows that normalization makes large parts of neural networks scale-invariant, creating a hidden feedback loop where learning-rate schedules and weight decay interact through the parameter norm to control the effective optimizer step. An exact discrete-time law with a single scalar quantity separates contraction- and expansion-dominated effective learning-rate regimes, and the balance point is intrinsically unstable, so constant learning rate with weight decay produces recurrent behavior instead of a stable equilibrium. A unified homogeneous-optimizer framework explains why adaptive methods stabilize more weakly under normalization. The law is validated with high precision on MLPs, CNNs, and GPT-2 across MNIST, CIFAR, WikiText, and OpenWebText, with code released on GitHub.

arXiv cs.AI / cs.LG / cs.CL·Sep 8, 17:46 UTC · 8d agoAI research

30

Previewing Ultrafast mode: GPT-5.6 Sol at up to 14X the speed

OpenAI previews Ultrafast, an API service tier running GPT-5.6 Sol up to 14x faster via Cerebras at up to 750 output tokens per second.

OpenAI announced a preview of Ultrafast, a new API service tier that runs GPT-5.6 Sol at up to 14 times the speed of standard inference. The tier is powered by Cerebras hardware and delivers up to 750 output tokens per second. The offering targets latency-sensitive developer workloads on OpenAI's API platform.

OpenAI News·Aug 13, 10:00 UTC · Aug 13, 2026AI tools & infra

45

ExecCritic: Learn to Test, Test to Improve for Coding Agents

ExecCritic separates test generation from patching for coding agents, lifting SWE-bench Verified resolution to 72.6%.

ExecCritic pairs a test-verify-revise scaffold with role-specific reinforcement learning: a Test agent writes repository-native tests and a Repair agent fixes code from execution feedback, both using Qwen-3.5-35B-A3B backbones. Post-trained Qwen agents compose to 72.6% on SWE-bench Verified, an 11.4-point gain over the 61.2% no-test baseline, without stronger-model or oracle feedback at evaluation time. The work shows test quality is the key variable: base-agent tests lowered resolution to 57.3% while GPT-5.6-sol tests raised it to 65.3%.

arXiv cs.AI / cs.LG / cs.CL·Sep 8, 17:53 UTC · 8d agoAI research1

36

TempCloze: Can Video-LLMs Identify the Missing Middle?

TempCloze benchmark tests Video-LLMs' temporal reasoning with 1,521 videos, finding temporal alignment is the primary failure mode across 31 models.

TempCloze is a video cloze benchmark in which models must identify the true missing middle clip given the beginning and ending clips, using 1,521 carefully filtered videos from seven sources, mostly long-take and egocentric footage. Distractors are constructed along three dimensions: Semantic, Alignment and Progression, with shared scenes and objects to reduce appearance cues. Evaluation of 10 proprietary and 21 open-source Video-LLMs found Alignment is the primary bottleneck, with models often recognizing plausible semantics and local event progression but struggling with temporal alignment.

Hugging Face daily papers·Aug 31, 20:00 UTC · 16d agoAI research

25

Verify Before You Distill: Prompt-Level Teacher Gating for On-Policy Distillation

TGOPD verifies teacher reliability per prompt before on-policy distillation, outperforming vanilla OPD across math, code, and instruction benchmarks.

Teacher-Gated On-Policy Distillation (TGOPD) estimates teacher reliability from verifier-scored teacher probes and routes each prompt either to dense on-policy distillation or to verifier-grounded GRPO, avoiding misleading updates from confidently wrong teachers under mode-seeking reverse KL. Across 4B and 35B students in mathematics, code, and instruction following, TGOPD outperforms vanilla OPD in all six single-domain settings and achieves higher seven-benchmark averages under multi-domain training. It also raises teacher-node GPU utilization from 9.8% to 78.9% in the measured 4B single-domain run by reusing idle teacher capacity.

Hugging Face daily papers·Sep 1, 20:00 UTC · 15d agoAI research

26

Testing race conditions with memory access tracing and stack-based delay injection

Google Project Zero released MAccConc, Linux kernel tooling that traces memory accesses to explore and test race condition interleavings.

A Google Project Zero researcher published MAccConc (Memory Access Concurrency), tooling for exploring possible interleavings of multithreaded test cases in the Linux kernel, available on GitHub. The tools use KCOV with ASAN outline-mode instrumentation to record per-access memory traces, enabling automatic testing of all A-B-A interleavings plus terminal and GUI explorers for manual analysis. The work targets confirming race condition candidates, building reliable regression tests, and enabling concurrency fuzzing, drawing on ideas from SKI and Ned Williamson's sockfuzzer.

Google Project Zero·Sep 8, 07:00 UTC · 8d agoResearch1

48

Fresh-Challenge VDF Attestations for Model-Relative Response Latency

Fresh-Challenge VDF Attestations bind verifiable delay functions to unpredictable public challenges, yielding succinct evidence of model-relative response latency.

The paper specifies FCLA, a protocol composition that binds a VDF to an unpredictable public challenge, a message, and independently auditable release/receipt records. Under explicit assumptions about VDF sequentiality and a calibrated bound on an adversary's sequential evaluation rate, an accepted transcript is inconsistent with post-challenge generation. A benchmark of the public reference implementation confirms the expected evaluation-versus-verification separation on one documented machine. The contribution is a protocol design analysis, not a new VDF construction.

arXiv cs.CR·Sep 11, 11:25 UTC · 5d agoResearch

15

Entropy-Regularized Rank-Masked Policy Optimization for Test-Time Reinforcement Learning in Code Generation

Researchers propose ERPO, enabling test-time reinforcement learning for code generation via probe-executed consensus rewards, rank masking, and entropy regularization.

The paper introduces probe-driven test-time reinforcement learning (TTRL) for code generation, where output-free probe inputs are constructed from problem statements and candidate programs are executed on them to compute a Probe Consensus Reward (PCR). Because PCR can be gamed through spurious consensus, the authors propose Entropy-Regularized Rank-Masked Policy Optimization (ERPO), which turns low-PCR outcomes into conservative negative updates via rank masking and constrains policy drift with an entropy ceiling. On coding benchmarks, ERPO substantially improves pass@1 and pass@k in both in-domain adaptation and zero-shot transfer.

arXiv cs.AI / cs.LG / cs.CL·Sep 8, 17:54 UTC · 8d agoAI research1

38

OpenAI’s GPT-5.6 Sol runs up to 14× faster with Ultrafast mode

OpenAI launched GPT-5.6 Sol Ultrafast mode in limited preview, running up to 14x faster at 750 tokens per second via Cerebras inference.

OpenAI's GPT-5.6 Sol Ultrafast mode is available in limited preview through the OpenAI API, delivering up to 14x faster processing and up to 750 output tokens per second, powered by Cerebras under the companies' ultra-low-latency inference partnership. Preview customers are testing it in production for coding, commerce, financial research, and support applications. OpenAI is also using Ultrafast internally for incident response tasks such as log analysis and trace review, and for research workflows with multiple same-day experiment iterations.

Help Net Security·Aug 14, 00:00 UTC · Aug 14, 2026AI industry

45

When the prompt becomes the payload: A practical pen-testing guide for GenAI, LLM and RAG applications

CSO Online publishes a practical penetration-testing guide for GenAI, LLM, and RAG applications, covering prompt injection, retrieval poisoning, and tenant isolation testing.

The guide frames LLM applications as attack graphs spanning prompts, retrieval layers, vector stores, tools, identities, and downstream APIs, arguing that conventional web testing misses instruction-vs-data channel risks. It builds on OWASP prompt injection guidance (direct vs. indirect injection) and NIST's 2025 adversarial machine-learning taxonomy, noting that RAG and fine-tuning do not remove injection risk. Recommended practices include documenting trust transitions across components, using canaries and synthetic records to avoid test side effects, running multi-turn and obfuscated injection campaigns, and verifying chains from poisoned documents to observable state changes. It also details testing RAG pipelines via controlled document poisoning across metadata, OCR layers, and code comments, plus cross-tenant isolation checks on retrieved document IDs.

CSO Online·Sep 9, 10:00 UTC · 7d agoAI safety & security1

28

DynSHAP: Towards Explainable Dynamic Survival Analysis

DynSHAP extends SHAP explainability to dynamic survival analysis, treating time-feature pairs as Shapley players for longitudinal clinical predictions.

DynSHAP adapts marginal SHAP estimators to dynamic survival analysis by treating time-feature pairs as players in the Shapley game, handling longitudinal irregular inputs and functional survival outputs. Temporal DynSHAP learns linear feature dependencies over time and addresses them with conditional sampling. On synthetic data with ground-truth attributions it recovers temporally dependent features more accurately than marginal estimators, and it produces faithful attributions on two real-world clinical datasets across two DSA architectures.

arXiv cs.AI / cs.LG / cs.CL·Sep 11, 16:39 UTC · 5d agoAI research

22

Show HN: LLM Attention Visualization

A developer released a browser-based tool that visualizes which past tokens influence each LLM output token using aggregated, value-weighted attention scores.

A Show HN project presents a React application built on Transformers.js that renders per-token attention influence by aggregating attention weights scaled by value-vector magnitudes across all attention heads and layers. To expose internal tensors, the author instrumented the ONNX computation graph, hosted a modified model on Hugging Face, and pre-generated prompts to avoid long model downloads in the browser. Demos with a 600-million-parameter model show how verbatim copying draws heavily on source tokens and how single outputs blend information from multiple phrases.

Hacker News · AI·Sep 8, 16:59 UTC · 8d agoAI tools & infraHN 41↑ · 9 comments

28

Your Critical Vulnerabilities Might Not Be Your Biggest Risk

Autonomous penetration testing advocates prioritize exploitable attack paths over raw vulnerability severity for continuous security validation.

The article argues that scanner severity scores lack context: a critical flaw behind strong segmentation may be low priority, while a medium flaw on internet-facing systems can provide a foothold chained toward sensitive data. It positions autonomous penetration testing and attack path validation as the execution layer for continuous security validation, replacing point-in-time assessments. The piece is vendor-authored thought leadership rather than incident or vulnerability news.

The Hacker News·Sep 11, 11:30 UTC · 5d agoIndustry1

18

Perturbation Probing: A New Diagnostic for the Fragility of LLM Safety

Unit 42 research shows LLM safety refusals concentrate in a thin neural layer, motivating external, multi-layered AI security controls.

Palo Alto Networks Unit 42 introduces Perturbation Probing, a diagnostic technique for measuring the fragility of LLM safety mechanisms. The research finds that safety refusal behavior is localized within a thin neural layer, implying small perturbations can undermine built-in refusals. The authors argue this motivates external, multi-layered security defenses on top of model-internal safety training.

Palo Alto Unit 42·Aug 28, 22:00 UTC · 18d agoAI safety & security

45

Likelihood-free inference with nuisance parameters through normalizing flows

Researchers decompose normalizing flows to derive near-pivotal statistics for likelihood-free inference with nuisance parameters, recovering the t-test and beating Welch limits.

A new paper decomposes neural-network normalizing flows to uncover pivotal statistics in the presence of nuisance parameters using only a sample generator from the distribution of interest. The statistic is near-pivotal in the sense of minimum average KL-divergence of its p-values and can incorporate prior knowledge of group invariances such as translation and scale. Experiments show it recovers the one-sample t-test almost exactly, outperforms the Welch test on worst-case size over a constrained variance-ratio range, and delivers higher power and much faster runtime than profile likelihood-ratio techniques on small-to-moderate samples.

arXiv cs.AI / cs.LG / cs.CL·Sep 9, 17:58 UTC · 7d agoAI research

16

TPMSpy: Validation of Measured Boot Systems by Low-Level Tracing of TPM Usage

Researchers present TPMSpy, a platform-agnostic method validating TPM Measured Boot via low-level tracing, finding inconsistent Linux systemd measurements that break remote attestation and LUKS decryption.

An arXiv paper (2609.05011) introduces TPMSpy, a method that analyzes virtualized system–TPM interactions to independently reconstruct and validate TPM Event Logs without relying on the quoting mechanism, applicable to open and closed systems and demonstrated on Linux and Windows. A longitudinal analysis of Linux systems running systemd versions 245–258 (2020–2025) found wide divergence in Measured Boot usage, undocumented behavioral changes, and no common usage pattern. The authors report inconsistent measurement of user-space systemd services, which prevents reliable remote attestation and LUKS disk decryption on affected systems.

arXiv cs.CR·Sep 4, 11:23 UTC · 12d agoResearch

40

Real-Time Intelligence with IBM Time Series Models on Confluent

IBM Research post on the Hugging Face blog describes running IBM time series models on Confluent for real-time intelligence.

Hugging Face's blog published an IBM Research post titled 'Real-Time Intelligence with IBM Time Series Models on Confluent.' No article body was available for classification, so details are limited to the title. The title indicates guidance on deploying IBM time series models alongside Confluent streaming infrastructure for real-time analytics.

Hugging Face Blog·Sep 2, 13:49 UTC · 14d agoAI tools & infra

14

What Else Needs Fixing? Exploring Cost-Effective Test-Time Compute for Revision Propagation in Artifacts Generated Through Conversation

A new benchmark shows LLMs reach 68.3-93% accuracy propagating local revisions across conversationally generated artifacts, with parallel-sample selection most cost-effective.

The paper introduces a benchmark for revision propagation: when users request a local change, LLMs must identify dependencies and update all affected parts of an artifact generated through conversation, where context lives in the chat history. Nine revision methods, including sequential reflection and parallel sampling variants, were evaluated on gpt-oss-20b/120b, gpt-5.4-mini, and qwen3.5-9b/27b/122b. Baselines scored 68.3-93% accuracy, and selecting among three parallel samples via LLM-based or medoid selection improved accuracy by 2.2-9.7% as the most cost-effective test-time compute strategy. Code and dataset are released.

Hugging Face daily papers·Sep 2, 20:00 UTC · 14d agoAI research

30

JustFit: 200K-Token LLM Serving on a 24 GiB Laptop with Just-in-Time State Management

JustFit MLX runtime serves 200K-token contexts for Qwen3.8-27B on a 24 GiB MacBook via just-in-time state management.

JustFit is an MLX-based inference runtime combining KVExec for compressed KV execution, PhaseSwap for component residency, and StateTrans for state-preserving serving transitions, independent of weight quantization. On a 24 GiB M4 Pro MacBook running Qwen3.8-27B MXFP4, it completed 196,608 input and 16,384 output tokens, raising single-request context from the mlx-vlm baseline's 30,720 positions to 212,992 (6.93x). Performance tests show 19.11 tokens/s on a 32K-input probe with a 16,374 MiB median peak footprint, and the runtime answered 29 of 30 AIME 2026 problems correctly.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 17:15 UTC · 1d agoAI research

38

"They don't care about this": A Systematic Study of TEE Build Reproducibility in the Wild

91% of 115 surveyed TEE deployments across Intel SGX, TDX, and AMD SEV fail to provide reproducible builds needed for verifiable remote attestation.

A systematic study of 115 TEE deployments found 91% were not reproducible and 80% lacked both source code and a reference build, undermining remote attestation guarantees. Interviews with 12 developers of 50 Intel SGX projects confirmed that only one participant treats reproducibility as a development priority. The authors identify technical barriers such as embedded timestamps plus ecosystem-level issues like lack of build-environment control in multi-stakeholder projects, and call for holistic, committed reproducibility practices.

arXiv cs.CR·Sep 10, 11:43 UTC · 6d agoResearch

38

PlannerForge: LLM Agents for Scenario-Based Testing of Motion Planners in Autonomous Driving

PlannerForge unifies scenario-based testing of autonomous driving motion planners in one LLM-agent framework, outperforming prior baselines.

PlannerForge is an LLM-agent framework that covers the full scenario-based testing pipeline for autonomous driving systems, spanning scenario generation, selection, modification, routing, planner testing, plus new enhancement and benchmarking stages. In evaluations with 10 off-the-shelf LLMs, best-per-task scores range from 0.88 to 1.00, and open-source 20-35B backends such as Qwen3.6:35B match commercial APIs on most tasks. End-to-end chaining retains 83% (commercial) and 78% (open) of seed queries, beats Scenario Factory 2.0 on executable generation, and cost-tuning lifts planner success from 50.4% to 70.2% while cutting collisions from 19.0% to 8.4%.

Hugging Face daily papers·Sep 7, 20:00 UTC · 9d agoAI research

28

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.

arXiv cs.AI / cs.LG / cs.CL·Sep 15, 16:29 UTC · 1d agoAI research

40

How well do agents use test/verification techniques?

Dan Luu's eval finds coding-agent testing instructions (TDD, formal methods, PBT, skills) mostly fail to beat defaults on Zstd implementation correctness.

The author ran 26 prompt conditions plus 4 skills on a Zstd-in-Rust implementation eval using codex with GPT-5.6, testing TDD, fuzzing, property-based testing, formal methods (Lean 4, TLA+, Verus, Kani, SMT solvers) and community skills. Nothing dramatically outperformed the default no-instruction condition, which did above average; at xhigh effort, fuzzing and PBT conditions did slightly better than formal methods. Pre-registered predictions included TDD underperforming and popular test skills (ECC, Hegel, Trail of Bits) not outperforming. Results are averages of 80 runs per condition plotted against cost.

Hacker News · security·Sep 8, 02:58 UTC · 8d agoAI researchHN 131↑ · 50 comments

38

Reference-Based Bias Detection in LLMs via Relative Representations of Hidden States

Researchers propose auditing LLM bias via relative hidden-state representations, detecting bias increases with 3-50x less compute than output-level benchmarks.

The paper introduces a reference-based bias auditing method that compares hidden-state representations across model variants, such as before and after fine-tuning, by encoding sentences relative to a fixed anchor set. The resulting Representational Bias Shift (Delta-B) correlates with output-level bias change in 15 of 18 tested settings, reaching |r| = 0.84 under full fine-tuning across WildGuardMix, DecodingTrust, and ToxiGen benchmarks. Thresholding Delta-B detects checkpoints whose bias increased with ROC AUC between 0.65 and 0.99 and beats a SEAT-based baseline, while auditing a model in about three minutes with 3-50x less compute.

Hugging Face daily papers·Sep 8, 20:00 UTC · 8d agoAI research1

42