← Back to all news
arXiv AI

Cortex: Measurement-Gated Evolution and Pruning of Coding-Agent Skills and Rules

Related stories

arXiv AI
Oct 9

Tracing the Thoughts of a Coding Agent Playing ARC-AGI-3: Lessons for Continual Learning

The paper investigates how a coding agent learns across a sequence of abstract reasoning tasks in the ARC-AGI-3 benchmark. The agent, operating on a frozen foundation model, writes and runs Python and shell scripts, retaining only its written artifacts as state. The authors develop a measurement protocol that traces each thought—any belief, rule, or plan committed to a file—through its lifecycle, revealing that most task-specific scripts are discarded at task boundaries and that the agent selectively forgets by rewriting knowledge into new scripts while accumulating unedited notes.

By Chen Wu, Josh Passenger, Yin Song
agents
More like this →
arXiv AI
Oct 9

One Skill Too Many: How Co-Installed Skills Conflict in Coding Agents

The paper investigates how co-installed coding‑agent skills can conflict, causing the model to choose a similar skill over the intended one based solely on name and description. An empirical study of 20,947 repositories uncovered 822,109 candidate skill pairs, with 312 confirmed conflicts tested across three models, revealing that nearly one in four installed skills is overridden by a similar skill. Key findings include that conflicts often involve normative skills, that the installed skill loses core functions in about one‑fifth of runs, and that the first skill read determines the outcome, suggesting benchmarks should evaluate exclusive core functions and platforms should guard the initial read.

By Chaoliang Yan, Zihao Xu, Yuekang Li, Shangzhi Xu, Yi Liu, Gelei Deng, Siqi Ma
llmsagentsbenchmarks
More like this →
arXiv AI
Sep 3

SkillGLoW: Procedural-Family Skill Consolidation for Self-Improving Agents on Long-Horizon Task Streams

SkillGLoW introduces a new way for large language model agents to self‑improve by consolidating procedural skills shared across related tasks. Instead of storing all skills in a single global document or a flat per‑task pool, SkillGLoW aggregates local skills into procedural families, compresses them into de‑instantiated global priors, and regenerates instance‑specific details on demand. Experiments on four diverse benchmarks show that these priors improve performance by an average of 17.2 points over a no‑skill baseline, are more compact than per‑task pools, and enable better transfer to unseen tasks.

By Ao Yan, Xin Zhang, Jiawei Du, Joey Tianyi Zhou
llmsagentsroboticsbenchmarks
More like this →
arXiv AI
Aug 19

StagedWorkspace: A Versioned Workspace for Knowledge-Work Agents

StagedWorkspace is a versioned workspace designed for knowledge‑work AI agents, ensuring that every parsed view, native file edit, and review diff is explicitly tied to a specific version of the workspace state. By binding parsed records and review diffs to content hashes of native files, the system improves performance on tasks such as OfficeQA and APEX‑Agents, achieving higher pass rates and rubric scores compared to single‑view approaches. The study demonstrates that providing dual parsed/native access and visible diffs enhances agent performance, highlighting workspace state as a key experimental variable for future benchmarks.

By Yining Hua, Hongbin Na, Yifan Zhou, Akshay Kalose, Cyrus Ayubcha, Levi Lian
llmsagentsbenchmarks
More like this →