Search: “Princeton University”

A misalignment of AI in mathematics

Fields Medalists including Terence Tao warn AI benchmark-chasing in mathematics is misaligned with scientific goals and threatens research culture.

Twenty-five Fields Medal-winning mathematicians, including Terence Tao, Peter Scholze, and Manjul Bhargava, published an essay arguing that AI companies' push to solve mathematical problems as benchmarks is severely misaligned with the goal of conceptual understanding. They warn that rushed AI solutions create attribution and plagiarism risks and bypass the human process of integrating results into the mathematical canon. The authors frame the issue as a broader alignment problem facing scientific and creative professions and call for urgent action by AI developers and society.

Hacker News · AI·Sep 11, 17:45 UTC · 5d agoAI safety & security 2 sourcesHN 59↑ · 153 comments

Import AI 469: Science AI; RSI simulator; and Zuck's technological pessimism

New DiG-bench benchmark of 70 hidden-rule games shows only Opus 5 and Fable 5 solving the hardest tiers, probing AI discovery and creativity.

Import AI 469 highlights DiG-bench (Discovery in Games), a benchmark of 70 handcrafted games with hidden rules and objectives where only 21 games are public and most are kept private to avoid training contamination. Only Opus 5 and Fable 5 with Claude Code solved any Tier 7 tasks (about 0.2 success), with GPT-5.5 next; the games are text-based and have beaten every human tester at least once. The newsletter also covers an RSI simulator game by Paradigm Research and Inherent's Faraday, a post-trained open-weight model that supervises frontier models to improve scientific research output.

Import AI·Aug 17, 13:05 UTC · Aug 17, 2026AI research

StudyBench: Can Self-Evolution Squeeze Textbooks for Olympiad Capability?

Researchers introduce StudyBench, a physics benchmark showing self-evolution gains on textbook problems rarely transfer to olympiad-level questions.

StudyBench is a controlled physics benchmark splitting test data into an Application Set of difficult textbook problems and a Transfer Set of olympiad-level problems. Across three base models, representative self-evolution methods improved on the Application Set but rarely transferred to the harder Transfer Set. A guidance ablation reveals a Guidance Gap, and every method hits a Compute Plateau, indicating the remaining limits are method problems rather than data or compute problems.

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

Crab Odyssey: Crustacean Wandered the Seas in a Bottle for Months, Scientists Find

404 Media·Aug 8, 10:00 UTC · Aug 8, 2026Industry

Unsolved Problem by Fields Medalist Breached by Two High School Students

Two high school students used Claude Opus 5 and GPT-5.6 Sol to help solve an open Lorentzian polynomials problem, posting a 75-page arXiv proof.

Aayush Bathija and Prince Rohatgi of Oak Park High School, mentored by UCLA postdoc Daniel Soskin, published the 75-page paper 'Bounded Ratios for Lorentzian Polynomials' (arXiv 2609.05341), solving an open problem in Fields Medalist June Huh's Lorentzian polynomial theory. The main structural theorem extends bounded coefficient-ratio characterization from quadratic to arbitrary-degree polynomials via discrete convexity conditions. The students used Claude Opus 5 and GPT-5.6 Sol for exploration and proof ideas but independently verified all arguments; the result follows an open letter from 25 Fields Medalists voicing concerns about AI's impact on mathematical rigor.

Hacker News · security·Sep 14, 12:23 UTC · 2d agoAI researchHN 22↑ · 19 comments

Nvidia CEO Jensen Huang tells Trump ‘we’re not going to let [an AI slowdown] happen’

Nvidia CEO Jensen Huang told Trump 'we're not going to let an AI slowdown happen,' rejecting Amodei's call to slow AI capability gains.

During the All-In Summit, Nvidia CEO Jensen Huang took a live speakerphone call from President Donald Trump and agreed 'we're not going to let' an AI slowdown happen, countering Anthropic CEO Dario Amodei's call to slow AI capability improvements, which Elon Musk and Sam Altman have endorsed. Trump called the slowdown sentiment 'a hoax' that could benefit political actors or China. Gallup polling shows seven in 10 Americans oppose local data center construction, with over 50% citing environmental resource concerns.

TechCrunch · AI·Sep 14, 21:51 UTC · 2d agoAI industry

Curriculum Learning as Transport: Understanding Curricula with Wasserstein Geodesics

Researchers model curriculum learning as Wasserstein transport over difficulty distributions, finding curriculum benefits are strongly task- and budget-dependent with no dominant strategy.

The framework represents curricula as trajectories of training distributions over discrete difficulty levels, decoupling ordering, matched exposure, endpoint smoothness, and pacing. Across a calibrated suite of 12 tasks and 33 difficulty axes under fixed training budgets, no single strategy dominates, though easy-to-hard ordering improves hard-level performance relative to exposure-matched static sampling. Endpoint smoothness and pacing substantially affect where along the difficulty spectrum a curriculum is effective, and the transport view supports extensions to learned pacing and structured difficulty spaces.

arXiv cs.AI / cs.LG / cs.CL·Sep 8, 17:39 UTC · 8d agoAI research1

A Princeton Researcher Proposes Recurrent Looped Transformer (RLT) that Carries Decoder State across Every Token, Fixing 96 Blocks per Token with Unbounded Temporal Depth

Princeton researcher Yifan Zhang proposes Recurrent Looped Transformer, carrying full decoder state across every token for unbounded temporal depth.

Yifan Zhang's technical report defines the Recurrent Looped Transformer (RLT), pairing a causal encoder with a recurrent decoder whose final output and layerwise sliding-window attention cache carry into every subsequent token with no prompt-response boundary reset. The reference configuration ties 48 encoder and 48 decoder layers, executing 96 logical blocks per token while the state path grows to 48t blocks after t tokens at fixed per-token compute. The report details RL replay contracts that rebuild all states under current parameters and exact prefix snapshots for multi-turn serving, but explicitly reports no measured efficiency, reasoning quality, or scaling results.

MarkTechPost·Sep 13, 17:02 UTC · 3d agoAI research1

RelateAnything: Real-Time Open-Vocabulary Relation Prediction From Any Inputs

RelateAnything is a 53M-parameter open-vocabulary relation prediction model running at 20 ms/frame, with 2.3-3.5x higher mean recall than comparable open-vocabulary methods.

RelateAnything predicts scored relations between image regions using any predicate vocabulary supplied at inference as text embeddings, with object labels never required as input, so region sources can change without retraining. Training covers 19,103 predicates using positive-unlabeled supervision; the authors release RA-4M (474k images, 4.3M geometrically verified relations over 10,102 free-text predicates) and the OV-SGG-Bench evaluation suite. The 53M-parameter model runs at 20 ms/frame and achieves 2.3-3.5x the mean recall of the strongest comparable open-vocabulary method across cross-dataset and zero-shot benchmarks. Model, corpus, and benchmark are public.

Hugging Face daily papers·Sep 10, 20:00 UTC · 6d agoAI research1

What must happen for AI’s trillion-dollar gamble to pay off

Hyperscalers need 2.7x productivity gains by 2030 to justify nearly $1.1 trillion in AI data center spending, or risk bankruptcy and capital misallocation.

Wharton finance professor Jessica Wachter estimates hyperscaler AI expenditure will reach nearly $1.1 trillion through 2027 and that a 2.7x productivity increase is needed to break even by 2030. AI revenues of roughly $150-200 billion this year fall far short of about $750 billion in annual spending, with total investment from Alphabet, Microsoft, Amazon, Meta, and Oracle potentially exceeding $5 trillion over four years. Alphabet reported its first free cash flow deficit (about $5.9 billion) since its 2004 IPO due to AI infrastructure costs. Researchers warn that failed demand could make the buildout the largest capital misallocation in history, with depreciating GPU chips risking stranded assets.

MIT Technology Review · AI·Sep 15, 10:00 UTC · 1d agoAI industry

One Symptom, Three Levers: A Critical Review of On-Policy Self-Distillation

A review paper frames on-policy self-distillation collapse as governed by three levers: token weighting, privileged information, and guidance decay.

The paper critically reviews On-Policy Self-Distillation (OPSD), where a language model trains on its own generations scored token-by-token by a teacher conditioned on privileged information such as reference solutions or environment feedback. It identifies collapse, the progressive narrowing of producible reasoning paths, as the dominant failure mode and analyzes it through three levers: signal weighting, the nature of privileged information, and teacher dynamics. The review is restricted to mathematical reasoning, reports no new experiments, and offers a shared vocabulary separating settled findings from disputed ones.

Hugging Face daily papers·Aug 25, 20:00 UTC · 22d agoAI research

HyQuant: Hybrid-Precision Quantization for LLM Attention

HyQuant keeps most LLM attention states low-bit while preserving vertical-line tokens and local windows in high precision, maintaining near-lossless accuracy.

HyQuant is a hybrid-precision quantization framework for LLM attention that quantizes most attention states to low bits while keeping accuracy-critical vertical-line tokens and local-window states in full precision, selected via lightweight attention-pattern signals. In the prefill stage it uses a hybrid-precision attention operator, and in the decode stage it applies the same principle to KV-cache compression with fused dequantization and attention computation. Across diverse tasks, models, and datasets it maintains nearly lossless accuracy; code is available on GitHub.

Hugging Face daily papers·Aug 27, 20:00 UTC · 20d agoAI tools & infra1

Necessary or Sufficient? Evaluating LLM Explanations With Behavioural Evidence

New framework tests whether LLM-cited explanation factors are necessary or sufficient, finding weak correlation across Claude, GPT, and Gemini models.

An arXiv paper introduces black-box intervention tests measuring whether factors LLMs cite in their explanations are necessary or sufficient for their outputs in agent oversight workflows. Across eight models from the Claude, GPT, and Gemini families, Spearman correlations between cited rankings and measured influence ranged from 0.349-0.354 (advisor recommendation) to 0.431-0.580 (prompt monitoring). Uncited factors scored above the lowest cited factor in up to 57.6% of advisor responses, showing cited top-three factors do not reliably identify the most influential inputs.

arXiv cs.AI / cs.LG / cs.CL·Sep 4, 17:37 UTC · 12d agoAI safety & security1

Verifiable by Construction: Claim-Level Evaluation of Verbatim Citation in Clinical Question Answering

Evaluation of twelve LLMs on 222 clinical questions shows verbatim quotes rarely substantiate claims; claude-opus-5 fully substantiates only 37.1%.

The authors build a standardized harness over four clinical practice guidelines and evaluate twelve LLMs on 222 synthetic clinical questions, measuring citation attachment, verbatim quote production, and claim substantiation. Most models attach verbatim quotes to over 90% of claims from prompting alone, though lightweight models like claude-haiku-4.5 struggle. Quotes frequently fail to substantiate claims: claude-opus-5 quotes 98.0% of claims but fully substantiates only 37.1%, exposing a capability gap for verifiable clinical QA.

arXiv cs.AI / cs.LG / cs.CL·Sep 14, 17:53 UTC · 2d agoAI research

I wrote an AI textbook — how long until AI can do it better?

AI researcher Nathan Lambert argues LLMs remain weak at long-form technical writing, questioning whether models can autonomously organize scientific knowledge for breakthroughs.

Nathan Lambert describes writing a post-training textbook, Reinforcement Learning from Human Feedback, and finds today's LLMs weak at organizing long-form technical content despite becoming superhuman at coding and math. He notes GPT 5.5 Pro found deep typos across a 200-300 page manuscript while Claude models proved more useful as editors. He argues that compressing knowledge through writing is a prerequisite for autonomous scientific insight and tempers expectations for near-term AI-driven open science.

Interconnects·Aug 12, 13:01 UTC · Aug 12, 2026AI research

Better answers, broader thinking: What students gain from ChatGPT and critical-thinking training

OpenAI published a randomized study of over 1,000 students examining ChatGPT's effects on critical thinking and performance.

OpenAI released results from a randomized study of more than 1,000 students using ChatGPT on a real-world university assignment. The research examined effects on critical thinking, originality, and student performance. Findings inform ongoing debates about AI's role in education and learning outcomes.

OpenAI News·Aug 27, 09:00 UTC · 20d agoAI research

Education Under Attack: The Pattern Behind Recent University Breaches

Huntress finds four recent university breaches share one root cause, security misconfigurations, and outlines fixes for higher education.

Huntress analyzed four university breaches from 2026 and identified misconfiguration as the common root cause behind the incidents. The report describes the recurring attack pattern targeting higher education and offers remediation guidance to close the gap. Specific victim institutions, threat actor attribution, and breach volumes are not named in the announcement.

Huntress·Aug 13, 15:00 UTC · Aug 13, 2026Threat actor in the wild

Structural Jailbreaks Generalize but Do Not Compound: A cross-provider and multilingual study of Involuntary In-Context Learning

Researchers show IICL structural jailbreaks generalize to Google Gemini, lifting attack success to 80-100% on harm and financial benchmarks; non-English prompts attenuate it.

The study red-teams two Google Gemini models with Involuntary In-Context Learning (IICL), a structural jailbreak reframing harmful requests as the final cell of a data-labeling task. IICL lifts attack success from at most 6.7% to 80-90% on HarmBench and 97-100% on financial abuse (FinProof), an order of magnitude above prior results on OpenAI's GPT-5.4. Against a compounding hypothesis, forcing IICL output into Spanish, Hindi, or Arabic attenuates the attack in 11 of 12 conditions, attributed to a 'relevance curse' producing lower-quality harmful content in lower-resource languages. Findings replicate under an independent non-Google judge (Cohen's kappa 0.86 over 377 paired verdicts).

arXiv cs.CR·Sep 8, 07:38 UTC · 8d agoAI safety & security

A*-Thought-V2: Efficient Latent Reasoning via Geometric Dynamics of LLM

A*-Thought-V2 compresses chain-of-thought into latent tokens using geometric hidden-state dynamics, cutting computation while improving accuracy on Qwen models.

A*-Thought-V2 models chain-of-thought as a hidden-state trajectory and interleaves explicit text with continuous latent tokens, compressing steps whose transitions deviate from the question-to-solution direction. Trained via stepwise embedding forcing and label forcing with soft multi-modal supervision, it was evaluated on Qwen3.5-9B and Qwen3.6-27B across six benchmarks. Reported results include up to 2.6% average accuracy gain, up to 50% shorter responses, 2.29x higher Accuracy per Computation Unit, 94.6% faster preprocessing, and up to 80.3% faster training.

Hugging Face daily papers·Sep 7, 13:56 UTC · 9d agoAI research

🔬“We have foundation models for language, not for physics” — Anima Anandkumar, Bren Professor of Computing

Caltech professor Anima Anandkumar discusses Neural Operators and FourCastNet for physics modeling, arguing inductive biases beat pure token scaling.

Anima Anandkumar, Bren Professor at Caltech and co-founder of Accelerated Understanding, describes Fourier Neural Operators that learn in frequency and spherical-harmonic domains to model weather, fusion, and fluid or heat flow. Her team built FourCastNet 3, a global weather model competitive with physics-based simulations that runs on consumer-grade GPUs. She also introduced TorchLean, a framework for writing PyTorch-style networks inside the Lean proof assistant for formal verification, and was appointed to the United Nations Scientific Advisory Board. She argues physical domains resist scaling due to tiny datasets and context lengths in the hundreds of billions, so progress comes from built-in structure and physical priors.

Latent Space·Aug 26, 15:15 UTC · 21d agoAI research1

The Mathematical AI Safety Institute wants to prove AI is safe the way cryptographers prove codes are unbreakable

Fields Medalist Jacob Tsimerman founds the Mathematical AI Safety Institute (MAISI) to pursue provable AI safety guarantees, launching January 2027.

Canadian mathematician and new Fields Medal recipient Jacob Tsimerman announced the Mathematical A.I. Safety Institute (MAISI), an independent Bay Area institute that will begin work in January 2027 with 10-30 mathematicians. Tsimerman is also joining OpenAI's safety team and argues AI needs a much higher safety standard. MAISI aims to formalize what safety means, prove multi-agent systems avoid unwanted outcomes, and explore tools like zero-knowledge proofs that let labs demonstrate correctness without exposing trade secrets.

The Decoder·Sep 11, 09:15 UTC · 5d agoAI safety & security

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.

arXiv cs.AI / cs.LG / cs.CL·Sep 4, 16:38 UTC · 12d agoAI research

Two-year university study finds banning AI from classrooms leaves students worse off

A two-year university study found students banned from using ChatGPT performed worst, while formal prompt-engineering training advantages faded as everyday AI familiarity grew.

Researcher Schrepel ran a classroom experiment in 2024 (66 students) and 2025 (164 participants) comparing a no-AI group, unguided ChatGPT users, and trained students revising EU AI Act provisions. The no-AI group finished last both years, hitting 'idea exhaustion' after 10-15 minutes, while the trained group's advantage nearly vanished by 2025 as everyday chatbot familiarity rose. Schrepel now argues blanket AI bans harm outcomes and universities should rethink bans and pure literature-review theses.

The Decoder·Sep 13, 09:27 UTC · 3d agoAI industry

The 2026 PNPL Competition: Word Classification and Efficient Cross-Subject Generalisation in LibriBrain100

The 2026 PNPL competition releases LibriBrain100, a MEG speech dataset with 32 extra subjects, targeting word classification and cross-subject BCI generalization.

The 2025 PNPL competition on non-invasive speech decoding from MEG achieved F1-macro scores of 95.6% for speech detection and 73.6% for phoneme classification, built on LibriBrain's ~50 hours of single-subject data. The 2026 edition extends this with LibriBrain100, adding 32 subjects (~40 minutes each) plus ~80 hours of within-subject data. Two tracks target within-subject word classification at scale and cross-subject generalization with subject-specific fine-tuning shrinking from ~40 to ~20 to ~10 minutes, aiming at clinically feasible non-invasive BCIs for people with profound paralysis.

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

Diffusion Models and Concept Formation

Paper argues diffusion models implicitly form Cobweb-like concept hierarchies, with a basic level emerging at intermediate noise levels.

The authors draw a formal correspondence between diffusion models and Cobweb, a classic incremental concept-hierarchy learner, noting both are hierarchical Bayesian density models with Gaussian prototypes. Modes of the diffusion model's noisy marginals form a hierarchy whose basic level sits at intermediate noise, where class identity commits. The correspondence is tested on MNIST and Fashion-MNIST via mode-finding. Diffusion is reframed as a cognitive model of concept formation.

arXiv cs.AI / cs.LG / cs.CL·Sep 11, 16:43 UTC · 5d agoAI research

Separating AI's Technological Problems from Its Capitalism Problems

Schneier and Sanders argue AI's harms stem from capitalist incentives and governance gaps, not just technical limits, urging structural reform.

Writing with Nathan E. Sanders in Tech Policy Press, Bruce Schneier argues that AI's technological problems (hallucination, sycophancy, overconfidence) must be separated from socio-political problems created by capitalist market incentives. The essay contrasts US frontier-scale, energy-intensive development with China's incentive-driven leaner open models on commodity hardware, and cites Switzerland's Apertus model - trained on licensed data using public computing and renewable hydropower - as a public-interest alternative. It contends that proposals like research pauses, data-center moratoria, and federal screening conflate technology problems with governance problems, and that society faces independent choices on both axes.

Schneier on Security·Aug 15, 00:00 UTC · Aug 15, 2026AI policy

IdeaAMBIG: Benchmarking Implementation-Critical Gaps in Research-Idea Specifications

IdeaAMBIG benchmark of 660 specification-gap instances shows LLMs localize implementation-critical research gaps poorly, with best model at 9.6% defect recovery.

IdeaAMBIG is a benchmark of 660 evidence-grounded instances evaluating whether research-method specifications provide enough information for faithful implementation: 163 real-world gaps from reproducibility reports and GitHub issues plus 497 controlled synthetic gaps. It tests codification-readiness assessment, defect localization, and clarification action generation across 13 LLMs. The best model achieves only a 9.6% Macro Defect Recovery Rate on real-world instances, though 80.6% clarification success when given the annotated defect, and an oracle study shows gold resolutions raise codification-ready rates from 14% to 98%. Defect localization emerges as the main bottleneck across all evaluated models.

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

A*-Thought-V2: Efficient Latent Reasoning via Geometric Dynamics of LLM

A*-Thought-V2 compresses redundant chain-of-thought steps into latent tokens guided by hidden-state geometry, improving accuracy up to 2.6% while halving response length.

A*-Thought-V2 models chain-of-thought as a hidden-state trajectory projected into a 3D PCA space and compresses steps whose transitions deviate from the question-to-solution direction into continuous latent tokens, keeping aligned steps explicit. Training uses stepwise embedding forcing and label forcing with soft multi-modal vocabulary supervision. On Qwen3.5-9B and Qwen3.6-27B across six in-domain and out-of-domain benchmarks it improves average accuracy by up to 2.6%, cuts response length by up to half, and raises Accuracy per Computation Unit 2.29x while reducing preprocessing and training time by 94.6% and up to 80.3%.

arXiv cs.AI / cs.LG / cs.CL·Sep 7, 17:56 UTC · 9d agoAI research

IdeaAMBIG: Benchmarking Implementation-Critical Gaps in Research-Idea Specifications

IdeaAMBIG benchmark with 660 instances measures whether LLMs can spot and fix underspecified research-method details for faithful implementation.

Researchers introduce IdeaAMBIG, a benchmark of 660 evidence-grounded instances (163 real-world gaps from reproducibility reports and GitHub issues, 497 controlled synthetic gaps) built from papers, codebases, and reproduction artifacts. It evaluates codification-readiness assessment, defect localization, and clarification action generation. Across 13 LLMs, the best model achieved only 9.6% Macro Defect Recovery Rate on real-world instances but 80.6% clarification success when given the annotated defect. An oracle study showed gold resolutions raise the codification-ready rate from 14% to 98%, identifying defect localization as the main bottleneck.

arXiv cs.AI / cs.LG / cs.CL·Sep 9, 17:59 UTC · 7d agoAI research2

Caltech Mathathon – first hackathon ever devoted to research level mathematics

Caltech will host the first research-level mathematics hackathon on October 30, giving 100 teams frontier AI models to attack open conjectures.

The Caltech Mathathon runs October 30 to November 1, assembling about 100 teams that will receive frontier models and over $2 million in AI credits to work on open mathematical problems. Teams will defend their results before leading mathematicians, with prize rounds before and after community verification of the results. The announcement cites recent AI-driven math results, including the disproof of Erdos's 80-year-old planar unit-distance conjecture, the first explicit non-sofic group, and a claimed complex structure on the six-sphere (unverified).

Hacker News · security·Sep 7, 09:26 UTC · 9d agoAI industryHN 264↑ · 92 comments1

Training and Finetuning Multi-Vector Embedding Models with Sentence Transformers

Hugging Face published a tutorial on training and finetuning multi-vector embedding models using the Sentence Transformers library.

Hugging Face's blog walks through training and finetuning multi-vector embedding models with Sentence Transformers. Multi-vector approaches store multiple vectors per document to support late-interaction retrieval. The post is a practical guide for developers building retrieval pipelines with the library.

Hugging Face Blog·Aug 26, 00:00 UTC · 22d agoAI tools & infra1

Multi-Vector (Late Interaction) Embedding Models with Sentence Transformers

Hugging Face details building and using multi-vector late-interaction embedding models with Sentence Transformers for retrieval workloads.

Hugging Face published a guide on multi-vector, late-interaction embedding models (ColBERT-style) supported through Sentence Transformers. The post covers how practitioners can build and use these models for retrieval and RAG pipelines. It is a developer tooling and technique write-up, not a security advisory.

Hugging Face Blog·Aug 18, 00:00 UTC · Aug 18, 2026AI tools & infra1

OpenAI’s feud with mathematicians is only escalating

25 Fields Medalists sign an open letter warning AI labs threaten math attribution; NYU's Tristan Buckmaster accuses OpenAI of pressuring him over collaborator credit.

Twenty-five Fields Medal-winning mathematicians signed an open letter arguing rushed AI proofs raise severe attribution and plagiarism questions and could destroy the culture of open research. NYU professor Tristan Buckmaster accused OpenAI of pressuring him not to credit an Anthropic-employed collaborator, and OpenAI withdrew sponsorship of a Caltech math event after researcher criticism. OpenAI's marathon-weekend proof remains unverified, and mathematicians fear their Codex usage may be fed into OpenAI's new models. The letter follows the June Leiden Declaration on LLM proofs.

TechCrunch · AI·updated Sep 12, 11:00 UTC · 4d ago·first Sep 11, 20:57 UTC · 5d agoAI industry 2 sources

K-Bench: a clinically calibrated benchmark for evaluating large language models in high-risk mental health conversations

Clinician-calibrated K-Bench evaluates 125 LLM configurations on 200 high-risk mental health vignettes, exposing wide variation in suicide and violence risk handling.

K-Bench is a clinician-calibrated, protected benchmark evaluating 125 model configurations from 33 base models across 14 providers on 200 multi-turn vignettes covering suicide, self-harm, domestic violence, substance misuse and no-risk presentations. A frozen GPT-4o judge achieved 94.2% exact agreement with clinician consensus across 6,751 eligible comparisons from 151 clinician-rated transcripts. Leading models combined supportive conversation with combined-risk scores above 95, while risk exploration varied substantially among weaker configurations; therapeutic prompting helped weaker models and elevated reasoning produced no average improvement. A continuously updated public leaderboard is hosted at k-bench.ai with protected test materials.

arXiv cs.AI / cs.LG / cs.CL·Sep 14, 16:49 UTC · 2d agoAI safety & security