Google's TEAL evolves reusable analytics tools for coding agents without ground-truth labels
google · hf · 2026-10-21
TEAL (Tool Evolution for Analytics over data Lakes) from Google, USC, and NYU tackles coding agents redoing data discovery from scratch for every question. Two agents propose tools from complementary evidence — an Explorer mining the data lake itself and a Critic analyzing Worker trajectories; a change is kept only if the Worker adopts the tool on a rerun without extra steps and any changed answer is backed by trajectory evidence. TEAL-Q synthesizes questions when real ones are unavailable. Across KramaBench and LakeQA with four Worker LLMs under tight 10- or 15-step budgets, TEAL improves accuracy in every setting.
More from coding & agent
- Claude Max now includes $100/month in API credits, screenshot hints at Opus 5.5 with 1M context — michalmalewicz · 2026-10-21
- Rejected 250 times in 2 years, this engineer landed a $500K offer after demoing 20 running agents — msharmas · 2026-10-21
- AI agent scores Steve Jobs' Stanford speech in GarageBand fully via computer use — itsOmSarraf_ · 2026-10-21
- Dev builds a four-agent pipeline for sponsored posts that adapts mid-campaign without restarting — alifcoder · 2026-10-21
- Ant International's FlowState treats execution state as memory for long-horizon LLM agents — ant-intl · 2026-10-21
- Tencent's ASAP evaluates agentic capabilities in pretrained base models before post-training — tencent · 2026-10-21