Frontend
Building verifiable databases at scale from the chemical literature via a multi-agent system
arXiv cs.MA0 views
arXiv:2507.20230v4 Announce Type: replace-cross Abstract: Accelerating chemical research with artificial intelligence requires high-quality databases. However, many reactions exist only as graphics drawn for human readers, which machines must read exactly. Specialist software is precise but rigid, whereas multimodal language models are flexible yet often write incorrect structures; neither flags errors. We present ChemEAGLE, a multi-agent system where tools supply structures and a language model extends their reach through judgments that programs apply. On 424 published graphics, its F1 score for exactly correct reactions was 82.2%, against 47.7% for a tool-equipped generalist agent, and calibrated confidence limited review to 35% of reactions at 96.6% accuracy. Without review, it built a 60,103-reaction photochemistry database whose records trained models to rank conditions and predict products. Verifiable databases can thus be built at scale.
Source:https://arxiv.org/abs/2507.20230PreviousEven Loose Proximity Constraints Make Makespan-optimal Unlabeled Multi-agent Pathfinding HardNextExplore, Map, Remember, Decide: A Benchmark for VLM Readiness in Safety-Critical Scenarios
Related posts
TLAPS-Bench: Formally Proving Correctness Of Real-World System Specs With TLA+
0Hacker News New
Show HN: csshm – cluster SSH written in Swift for macOS
0Hacker News New
Alaska Senate race called: Democrats win
0Hacker News New
Don't ask the model to be brief. Cut it in code, and count what you cut.
0dev.to

Sutton Hoo's Garnets Glow Because of the Gold Beneath Them
0dev.to

How to Add Google Analytics to WordPress (Step by Step)
0dev.to Beginners