
The ICML 2026 Agent Reproduction Audit: 23% of Examined Papers Had Falsified or Contested Claims
ICML 2026 reproduction audit: coding agents checked 2,226 papers claim by claim, and 23% had a falsified or contested claim. What the challenge found.
424 articles - page 10 of 36

ICML 2026 reproduction audit: coding agents checked 2,226 papers claim by claim, and 23% had a falsified or contested claim. What the challenge found.

LLM judges flip verdicts under pushback and rankings shift with token budget. The design spec for a judge system, including cheap deliberating juries.

AutoGPT's playbook for AI pull requests: AGENTS.md placement, enforced PR templates, trigger-phrase skills and a CLA that keep agent-heavy repos reviewable.

DeepSeek Harness (dsh) code read: a 453K-line plugin runtime on a vendored Cordis fork, three patterns worth stealing, and the gaps the launch does not mention.

A new August 2026 paper argues that coding agents improve faster when they compare attempts across tasks and lineages, not just retry one failed trajectory.

CLAUDE.md and AGENTS.md files keep growing because the reasoning behind each rule decays. A study of 1,867 repos names the mechanism and a comment-based fix.

Dub videos into other languages with the ElevenLabs Dubbing API and a coding agent that extracts product names, keeping speakers, timing and background audio.

EvoX Genesis built a Rust C compiler for $44 in tokens by keeping the project persistent and agents finite-lived. What the design says about agent memory.

OpenAI Enterprise Signals data: Codex drives 64% of enterprise output tokens and frontier firms generate 8.3x the tokens per user of typical firms.

Skill files are a supply chain attack surface: adversarial skills exploit a mainstream coding agent in over 95% of runs. The fix is admission engineering.

ACE vs ALTK-Evolve for agent memory: both turn trajectories into lessons, but calibrated delivery matches accuracy on AppWorld at a fraction of the tokens.

Cactus open-sourced Needle 2, a 45M-parameter agentic LLM in a single 14MB binary that runs a full tool-calling session in 28MB of RAM. 500 tok/s on a Raspberry Pi 5, ESP32-S3 class parts, Apache 2.0. Here is what the benchmarks actually show.
Showing 109-120 of 423 articles

New tutorials, open-source projects, and deep dives on coding agents - delivered weekly.
Explore 841 topics
Browse All Topics