Search: “open-weights”

AgentAtlas: Beyond Outcome Leaderboards for LLM Agents

arXiv:2605.20530v1 Announce Type: cross Abstract: Large language model agents now act on codebases, browsers, operating systems, calendars, files, and tool ecosystems, but the benchmarks used to evalu

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Do Vision--Language Models Understand 3D Scenes or Just Catalogue Objects?

DGX agent

arXiv:2605.20448v1 Announce Type: new Abstract: Vision--language models reliably name objects in a scene, but do they represent the 3D layout those objects inhabit? We introduce a 3,034-sample human-c

model-releasesarxiv-cs-cv

21 May 2026

Reasoning-Trace Collapse: Evaluating the Loss of Explicit Reasoning During Fine-Tuning

arXiv:2605.21127v1 Announce Type: new Abstract: Explicit reasoning models are trained to produce intermediate reasoning traces before final answers, but downstream fine-tuning is often performed on or

researcharxiv-cs-lg

SCRIBE: Diagnostic Evaluation and Rich Transcription Models for Indic ASR

arXiv:2605.20712v1 Announce Type: new Abstract: Automatic speech recognition replaces typing only when correction costs less than manual entry, a threshold determined by error types, not counts: fixin

safetyarxiv-cs-cl

TASTE: A Designer-Annotated Multi-Dimensional Preference Dataset for AI-Generated Graphic Design

model-releasesarxiv-cs-cv

arXiv:2605.20731v1 Announce Type: new Abstract: Text-to-image models produce graphic design at production scale, but their supervision comes from photo-style preference data with a single overall verd

TempGlitch: Evaluating Vision-Language Models for Temporal Glitch Detection in Gameplay Videos

model-releasesarxiv-cs-cv

arXiv:2605.21443v1 Announce Type: new Abstract: Vision-language models (VLMs) are increasingly being explored for video game quality assurance, especially gameplay glitch detection. Most existing eval

Text Analytics Evaluation Framework: A Case Study on LLMs and Social Media

arXiv:2605.21338v1 Announce Type: new Abstract: LLMs have demonstrated exceptional proficiency in a wide range of NLP tasks. However, a notable gap remains in practical data analysis scenarios, partic

torchtune: PyTorch native post-training library

arXiv:2605.21442v1 Announce Type: new Abstract: Modern LLMs typically require multistage training pipelines to achieve strong downstream performance, with post-training serving as the main interface f

researcharxiv-cs-lg

Detecting Fluent Optimization-Based Adversarial Prompts via Sequential Entropy Changes

arXiv:2605.19966v1 Announce Type: cross Abstract: Optimization-based adversarial suffixes can jailbreak aligned large language models (LLMs) while remaining fluent, weakening static and windowed perpl

From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models

arXiv:2605.20177v1 Announce Type: new Abstract: Recent advances in vision-language models (VLMs) emphasize long chain-of-thought reasoning; yet, we find that their performance on visual tasks is prima

researcharxiv-cs-cl

HalluWorld: A Controlled Benchmark for Hallucination via Reference World Models

arXiv:2605.19341v1 Announce Type: cross Abstract: Hallucination remains a central failure mode of large language models, but existing benchmarks operationalize it inconsistently across summarization,

Lying Is Just a Phase: The Hidden Alignment Transition in Language Model Scaling

arXiv:2605.18838v1 Announce Type: cross Abstract: Scaling laws predict loss from compute but not how capabilities interact. We measure the coupling between reasoning and truthfulness across 63 base mo

POLAR-Bench: A Diagnostic Benchmark for Privacy-Utility Trade-offs in LLM Agents

arXiv:2605.19127v1 Announce Type: new Abstract: LLM agents increasingly have access to private user data and act on the user's behalf when interacting with third-party systems. The user defines what m

The Growing Pains of Frontier Models: When Leaderboards Stop Separating and What to Measure Next

arXiv:2605.18840v1 Announce Type: cross Abstract: Leaderboards rank frontier models on independent axes but do not reveal whether capabilities reinforce or trade off across releases -- and at the fron

The Silent Hyperparameter: Quantifying the Impact of Inference Backends on LLM Reproducibility

model-releasesarxiv-cs-lg

arXiv:2605.19537v1 Announce Type: new Abstract: Progress in LLMs is increasingly measured through standardized benchmarks, where state-of-the-art improvements are often separated by fractions of a per

Toto 2.0: Time Series Forecasting Enters the Scaling Era

arXiv:2605.20119v1 Announce Type: cross Abstract: We show that time series foundation models scale: a single training recipe produces reliable forecast-quality improvements from 4M to 2.5B parameters.

Beyond Accuracy: Decomposing the Reasoning Efficiency of LLMs

arXiv:2602.09805v2 Announce Type: replace-cross Abstract: As reasoning LLMs increasingly trade tokens for accuracy through deliberation, search, and self-correction, a single accuracy score can no lon

Applications

Beyond Sentiment Classification: A Generative Framework for Emotion Intensity Evaluation in Text

arXiv:2605.16613v1 Announce Type: new Abstract: We introduce a novel approach to emotion modeling that shifts the focus from identification to evaluation, addressing the limitations of discrete classi

applicationsarxiv-cs-cl

ChemVA: Advancing Large Language Models on Chemical Reaction Diagrams Understanding

arXiv:2605.17214v1 Announce Type: new Abstract: While Large Language Models (LLMs) have revolutionized scientific text processing, they exhibit a significant capability gap when interpreting chemical

safetyarxiv-cs-ai

DISA: Offline Importance Sampling for Distribution-Matching LLM-RL

arXiv:2605.17295v1 Announce Type: cross Abstract: Modern reasoning agents are increasingly evaluated on their ability to generate multiple valid solution paths, plans, or tool-use traces for a given i

safetyarxiv-cs-cl

HydroAgent: Closing the Gap Between Frontier LLMs and Human Experts in Hydrologic Model Calibration via Simulator-Grounded RL

model-releasesarxiv-cs-lg

arXiv:2605.17792v1 Announce Type: new Abstract: Calibrating distributed hydrologic models is a critical bottleneck across operational water resources management - streamflow prediction, reservoir oper

LEAF: A Living Benchmark for Event-Augmented Forecasting

arXiv:2605.16358v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly applied to forecasting. To evaluate this capability while mitigating pre-training data contamination, se

Agents

Lying with Truths: Open-Channel Multi-Agent Collusion for Belief Manipulation via Generative Montage

arXiv:2601.01685v2 Announce Type: replace-cross Abstract: As large language models (LLMs) transition to autonomous agents synthesizing real-time information, their reasoning capabilities introduce an

agentsarxiv-cs-ai

Meltdown: Circuits and Bifurcations in Point-Cloud-Conditioned 3D Diffusion Transformers

arXiv:2602.11130v2 Announce Type: replace-cross Abstract: Sparse point clouds are a common input modality for 3D surface reconstruction, including in safety-critical settings such as surgical navigati

safetyarxiv-cs-cv

NewsLens: A Multi-Agent Framework for Adversarial News Bias Navigation

arXiv:2605.17364v1 Announce Type: new Abstract: Media bias detection has predominantly been framed as a classification task: assign a political label to an article or outlet. We argue this framing is

Agents

OProver: A Unified Framework for Agentic Formal Theorem Proving

arXiv:2605.17283v1 Announce Type: cross Abstract: Recent progress in formal theorem proving has benefited from large-scale proof generation and verifier-aware training, but agentic proving is rarely i

agentsarxiv-cs-ai

ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge

arXiv:2510.18941v2 Announce Type: replace-cross Abstract: Evaluating progress in large language models (LLMs) is often constrained by the challenge of verifying responses, limiting assessments to task

SAME: A Semantically-Aligned Music Autoencoder

arXiv:2605.18613v1 Announce Type: cross Abstract: Latent representations are at the heart of the majority of modern generative models. In the audio domain they are typically produced by a neural-audio

Sustainability via LLM Right-sizing

arXiv:2504.13217v3 Announce Type: replace-cross Abstract: Large language models (LLMs) have become increasingly embedded in organizational workflows. This has raised concerns over their energy consump

Test-Time Hinting for Black-Box Vision-Language Models

arXiv:2605.16410v1 Announce Type: new Abstract: Test-time scaling (TTS) methods have proven highly effective for LLMs, yet their application to vision-language models (VLMs) remains relatively underex

researcharxiv-cs-cv

Applications

Traces of Social Competence in Large Language Models

arXiv:2603.04161v2 Announce Type: replace Abstract: The False Belief Test (FBT) has been the main method for assessing Theory of Mind (ToM) and related socio-cognitive competencies. For Large Language

applicationsarxiv-cs-cl

Fair outputs, Biased Internals: Causal Potency and Asymmetry of Latent Bias in LLMs for High-Stakes Decisions

arXiv:2605.15217v1 Announce Type: new Abstract: Instruction-tuned language models exhibit behavioural fairness in high-stakes decisions while retaining biased associations in their internal representa

Fully Open Meditron: An Auditable Pipeline for Clinical LLMs

arXiv:2605.16215v1 Announce Type: new Abstract: Clinical decision support systems (CDSS) require scrutable, auditable pipelines that enable rigorous, reproducible validation. Yet current LLM-based CDS

How to Train Your Advisor: Steering Black-Box LLMs with Advisor Models

arXiv:2510.02453v3 Announce Type: replace-cross Abstract: Frontier language models are deployed as black-box services, where model weights cannot be modified and customization is limited to prompting.

Judge Circuits

arXiv:2605.16023v1 Announce Type: new Abstract: LLM-as-a-judge has become the dominant paradigm for grading model outputs at scale, yet the same model assigns systematically different scores when its

Measuring Maximum Activations in Open Large Language Models

arXiv:2605.15572v1 Announce Type: new Abstract: The dynamic range of activations is a first-order constraint for low-bit quantization, activation scaling, and stable LLM inference. Prior work characte

Painless Activation Steering: An Automated, Lightweight Approach for Post-Training Large Language Models

arXiv:2509.22739v3 Announce Type: replace-cross Abstract: Language models (LMs) are typically post-trained for desired capabilities and behaviors via weight-based or prompt-based steering, but the for

We built TERMS-Bench, a three-tier benchmark for LLM agents in real-world economic negotiation. No LLM-as-judge, no outcome rubrics: the env…

model-releaseszhipu-ai--x

We built TERMS-Bench, a three-tier benchmark for LLM agents in real-world economic negotiation. No LLM-as-judge, no outcome rubrics: the environment itself is the verifier. 🏆Among frontier models, @An

17 May 2026

A Deterministic Agentic Workflow for HS Tariff Classification: Multi-Dimensional Rule Reasoning with Interpretable Decisions

arXiv:2605.14857v1 Announce Type: new Abstract: Harmonized System (HS) tariff classification is a high-stakes, expert-level task in which a free-form product description must be mapped to a specific s

CurveBench: A Benchmark for Exact Topological Reasoning over Nested Jordan Curves

model-releasesarxiv-cs-cv

arXiv:2605.14068v1 Announce Type: new Abstract: We introduce CurveBench, a benchmark for hierarchical topological reasoning from visual input. CurveBench consists of extbf{756 images} of pairwise non-

Do We Really Need External Tools to Mitigate Hallucinations? SIRA: Shared-Prefix Internal Reconstruction of Attribution

arXiv:2605.14621v1 Announce Type: cross Abstract: Large vision-language models (LVLMs) often hallucinate when language priors dominate weak or ambiguous visual evidence. Existing contrastive decoding

researcharxiv-cs-ai

HodgeCover: Higher-Order Topological Coverage Drives Compression of Sparse Mixture-of-Experts

arXiv:2605.13997v1 Announce Type: cross Abstract: Sparse Mixture-of-Experts (MoE) layers route tokens through a handful of experts, and learning-free compression of these layers reduces inference cost

researcharxiv-cs-ai

Agents

Hypergraph Enterprise Agentic Reasoner over Heterogeneous Business Systems

arXiv:2605.14259v1 Announce Type: new Abstract: Applying Large Language Models (LLMs) to heterogeneous enterprise systems is hindered by hallucinations and failures in multi-hop, n-ary reasoning. Exis

agentsarxiv-cs-ai

Mining Subscenario Refactoring Opportunities in Behaviour-Driven Software Test Suites: ML Classifiers and LLM-Judge Baselines

arXiv:2605.14568v1 Announce Type: cross Abstract: Context. Behaviour-Driven Development (BDD) software test suites accumulate duplicated step subsequences. Three published refactoring patterns are ava

ML-Embed: Inclusive and Efficient Embeddings for a Multilingual World

arXiv:2605.15081v1 Announce Type: cross Abstract: The development of high-quality text embeddings is increasingly drifting toward an exclusionary future, defined by three critical barriers: prohibitiv

SciPaths: Forecasting Pathways to Scientific Discovery

arXiv:2605.14600v1 Announce Type: new Abstract: Scientific progress depends on sequences of enabling contributions, yet existing AI4Science benchmarks largely focus on citation prediction, literature

When Evidence Conflicts: Uncertainty and Order Effects in Retrieval-Augmented Biomedical Question Answering

arXiv:2605.14115v1 Announce Type: new Abstract: Biomedical retrieval-augmented large language models (LLMs) often face evidence that is incomplete, misleading, or internally contradictory, yet evaluat

researcharxiv-cs-cl

LLMs as annotators of credibility assessment in Danish asylum decisions: evaluating classification performance and errors beyond aggregated metrics

arXiv:2605.13412v1 Announce Type: cross Abstract: Off-the-shelf large language models (LLMs) are increasingly used to automate text annotation, yet their effectiveness remains underexplored for underr