verification

This article introduces SkillGen, a multi-agent framework that synthesizes and verifies reusable inference-time skills for LLM agents by contrasting successful and failed trajectories. The method ensures skills are auditable and empirically verified for their net positive impact on agent performance.

DeltaRubric is a research paper introducing a two-step multimodal preference evaluation approach using a single MLLM to improve reward modeling reliability through joint planning and verification.

Google's next-generation reCAPTCHA now requires Play Services on Android, breaking verification for de-Googled users and raising privacy concerns about ecosystem control.

This paper introduces TGS-RAG, a bidirectional verification and completion framework that synergizes text-based and graph-based Retrieval-Augmented Generation to improve multi-hop reasoning accuracy.

The article discusses the importance of quality control for reinforcement learning data, outlining the shortcomings of current data vendors and the evaluation criteria used by frontier AI labs for RL data.

A critical blog post argues Anthropic's claims about Claude Mythos finding thousands of zero-days are unsubstantiated, noting the 244-page system card lacks CVEs, CVSS scores, or independent verification, undermining trust in the model's safety narrative.

Researchers release λ-RLM, an open-source typed λ-calculus runtime that replaces self-written recursive control code with pre-verified combinators, boosting long-context reasoning accuracy by up to 21.9% and winning 29/36 trials.

C2 proposes a scalable rubric-augmented reward modeling framework that trains a cooperative rubric generator and critical verifier exclusively from binary preferences, eliminating the need for costly rubric annotations while achieving up to 6.5 point gains on RM-Bench.

OpenAI researchers found that optimizing language models purely for correct answers reduces human interpretability, and propose 'prover-verifier games' where a prover generates solutions and a verifier checks them, improving legibility for both humans and AI systems.