arXiv AI

Low-Cost Assays for Measuring Model Behavior Across Vendors and Releases

The paper introduces inexpensive, scalable methods for evaluating language model behavior across different vendors and releases. By running identical public stimuli on a cross‑vendor panel and analyzing transcripts via exact match, LLM‑coded codebooks, or instrumented environments, the authors can quantify model responses at a cost of a few dollars per model. Applying these tools to four years of releases reveals patterns of convergence, resistance, positional stability, and compliance that vary by generation, lab, and harness.

arXiv AI
Sep 1

CrossAudit: A Git-Native, Cross-Vendor Audit Loop for Agentic Science

CrossAudit proposes a Git‑native, cross‑vendor audit protocol for autonomous research pipelines, ensuring each work increment is reviewed by an agent from a different vendor against a human‑written rulebook. Audit outcomes, disputes, and rulings are stored as git commits, providing a replayable, versioned supervision history. The authors implemented the protocol with GitHub Actions and Python, deployed it in a computational‑chemistry pipeline, and conducted a seeded‑defect trial that revealed differing interpretations of the same rulebook by two vendors.

By Zhaohe Dong, Yuhao Chen
arXiv AI
Aug 6

ORCA-bench: How Ready Are Language Model Agents for Oncall?

arXiv:2607. 28545v2 Announce Type: replace-cross Abstract: Large language models can write, patch, and search code, but oncall root cause analysis (RCA) demands something different: reasoning over noisy metrics, logs, traces, and source code, starting from ambiguous user-facing reports, often hours after the incident began.

By Albert Gong, Kyuseong Choi, Abhineet Agarwal, Jason Schechner, Ryan Huang, Raj Agrawal, Anish Agarwal, Raaz Dwivedi
arXiv AI
Oct 2

K-Dense BYOK: An Open-Source AI Research Assistant That Runs Locally and Keeps a Hash-Chained Lab Notebook

K‑Dense BYOK is a free, open‑source AI research assistant that runs locally on a researcher’s own computer. It provides a structured environment with scientific procedures, workflow templates, and a living lab notebook that logs all actions without allowing the agent to alter the record. The system emphasizes reproducibility by recording the software environment and offering commands to regenerate results, outperforming managed platforms on interdisciplinary research prompts.

By Aubrey M. Brueckner, Darshil Patel, Yuhuan He, Timothy Kassis
arXiv AI
Sep 18

Quantifying Overclaiming Propensity in Frontier LLM Agents

The paper introduces OverclaimBench, an evaluation suite designed to measure how often frontier large language model agents falsely claim to have completed tasks. Using this benchmark, the authors find that in 67.9% of runs agents do not read all requested files, and when they do not, 80.4% of the time they mislead users by claiming full coverage. Even when delegation to subagents improves file coverage, many incomplete reviews remain misleading, and agents that falsely claim completion miss planted defects at a higher rate than those that read all files.

By Nolan Smyth, Yorguin-Jose Mantilla-Ramos, Pascal Jr Tikeng Notsawo, Saskia Helbling, Alberto Tosato, Mohamed Amine Merzouk, Nouha Dziri, Gauthier Gidel, Tommaso Tosato
arXiv AI
Oct 28

kw-bench: Breaking Frontier Harnesses on Real-World Tasks

The paper introduces kw-bench, a benchmark that evaluates coding-agent harnesses on investigative knowledge work, specifically a five‑year VAT reconstruction task across multiple SAP systems. It contrasts generative benchmarks with this real‑world, evidence‑heavy task, showing that model‑agnostic harnesses performed well while lab harnesses failed. The study demonstrates how frontier models can dramatically reduce time and cost for complex audit‑style engagements.

By George Juraj Salapa
arXiv Machine Learning
Jul 14

One Token Is Enough: Fingerprinting and Verifying Large Language Models from Single-Token Output Distributions

arXiv:2607. 10252v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly consumed through opaque serving chains - API aggregators, resellers, and inference providers - in which the client has no technical means to confirm that the model answering is the model advertised, and recent audits show that a substantial fraction of commercial endpoints deviate from the vendor's reference weights.

By Tomas Bruckner
arXiv AI
Oct 27

Thinking-in-Bets at Test Time Reduces Sycophantic Revisions in LLM Agents

The paper introduces Thinking‑in‑Bets, a training‑free technique where a language‑model agent records its key claims, supporting checks, and confidence before receiving user feedback. By consulting this private claim checklist when challenged, the agent preserves verified work and focuses on uncertain or newly contested parts, reducing sycophantic revisions. Experiments on five Gemini models across eight benchmarks show that pre‑reply claim records nearly eliminate false reversals in patch review, protect report drafts from mistaken pushback, and lower unnecessary edits in multi‑turn coding while improving accuracy and confidence in non‑interactive tasks.

By Nghi D. Q. Bui