On standard factuality tasks, frontier models now cluster near the top of the scale. The question is therefore shifting from how factual a system is toward how much compute that factuality costs.
arXiv:2607. 24063v1 Announce Type: new Abstract: On standard factuality tasks, frontier models now cluster near the top of the scale.
Agent evaluations increasingly benchmark LLMs, but rankings can be swayed by evaluation conditions such as scaffolds or tasks, making reliability claim‑dependent. A Bayesian variance‑decomposition framework applied to 22 benchmarks shows that reliability varies with the measurement goal: fixed model‑scaffold systems rank reliably, while underlying‑model rankings are less stable. Scaffold choice can alter conclusions, and adding more tasks only modestly improves reliability when scaffold coverage is limited; however, pooling diverse benchmarks can substantially raise cross‑task ranking reliability and reduce cost.
AgentJudgeBench is a new benchmark that evaluates the reliability of large language model (LLM) judges on agentic tool‑calling tasks involving workflow directed acyclic graphs (DAGs). It contains 3,808 instances across six DAG topologies and three difficulty tiers, tested with five generators (3B–70B open‑weight models and GPT‑5.4) and six judges (20B to frontier scale) under both paired‑with‑and‑without‑ground‑truth conditions. The study finds that judge alignment degrades with task difficulty, ground‑truth exposure can sometimes hurt alignment, and structured evaluation rubrics provide modest improvements, revealing a structural ceiling that model capacity alone cannot surpass.
arXiv:2606. 08151v1 Announce Type: new Abstract: Tool-using LLM agents often fail not because relevant text is absent, but because decisive evidence is not selected, compressed, or surfaced at action time.
arXiv:2609.07785v1 Announce Type: new Abstract: An LLM-agent leaderboard invites a familiar inference: an agent ranked above another is the better agent. Public evaluation logs may not support that c...
arXiv:2606. 21008v2 Announce Type: replace-cross Abstract: The metanym game is a competitive word game for LLMs that measures structural intelligence against established cognitive-science constructs.
arXiv:2608. 00004v1 Announce Type: cross Abstract: Grading natural-language mathematical proofs is a recurring cost in evaluating math-reasoning systems, and frontier LLM judges are expensive.
arXiv:2607. 02104v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used as cheap, scalable judges that compare candidate outputs pairwise -- to rank responses, select models, or triage papers.
arXiv:2607. 01211v1 Announce Type: cross Abstract: Repository-level performance-optimization benchmarks such as GSO, SWE-Perf and SWE-fficiency evaluate coding agents by applying patches to real repositories and comparing runtime against unoptimized baselines and official reference patches.
arXiv:2606. 03650v1 Announce Type: cross Abstract: Choosing or ranking language models for a specific application is hardest when no task-specific labeled data exists, and standard public benchmarks cannot be trusted, their items having likely leaked into pretraining, so scores reflect memorization rather than fitness.
arXiv:2607. 07946v1 Announce Type: cross Abstract: DeepSWE is a benchmark of 113 original, long-horizon software engineering tasks for evaluating coding agents.
The paper introduces a penalty‑aware evaluation framework for Retrieval‑Augmented Generation (RAG) systems that uses asymmetric scoring, knowledge‑gap canaries, and a failure‑attribution pipeline. Applying this framework to three commercial RAG products and a baseline on SimpleQA‑Verified, the authors find that while overall accuracy is similar across systems, canary violation rates vary dramatically, showing that systems differ more in when they answer than in what they answer. The study demonstrates that penalty‑aware scoring can reorder system rankings and is robust across different penalty settings.