arXiv AI
TasteVal: Measuring the Experimental Research Taste of AI Systems Against Human Experts
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Sep 24
The Tasteful Agent: Measuring and Improving Taste in Long-Horizon Tasks
The paper introduces the concept of an agent’s "taste"—its ability to make effective long‑horizon decisions—and presents Taste‑Bench, a new benchmark that automatically generates decision‑fork questions from agent trajectories. Taste‑Bench evaluates models on choosing the best path without seeing future outcomes, revealing that top models answer only about 60% of questions correctly and that later‑appearing evidence makes forks harder. The authors also demonstrate that training a student model to mimic a teacher’s judgment improves decision quality and overall success on held‑out software engineering tasks.
arXiv AI
Jul 13
Long-Horizon-Terminal-Bench: Testing the Limits of Agents on Long-Horizon Terminal Tasks with Dense Reward-Based Grading
arXiv:2607. 08964v1 Announce Type: new Abstract: AI agents have become capable of autonomously completing short, well-specified tasks.
arXiv AI
Aug 17
AI Research Preference Models
arXiv:2608. 13940v1 Announce Type: new Abstract: AI research agents (AIRA) can now propose, implement, and evaluate their own machine learning experiments, but progress on frontier tasks is throttled by cost: a candidate solution can be written in minutes, whereas evaluating it can take hours to days of GPU time.
arXiv AI
Jun 3
A Matter of TASTE: Improving Coverage and Difficulty of Agent Benchmarks
arXiv:2605. 28556v2 Announce Type: replace Abstract: As agent capabilities advance, existing benchmarks, such as $\tau^2$-Bench, are becoming increasingly saturated.
arXiv AI
Sep 11
OpenDiscoveryTrace: Process Traces for Evaluating AI Scientist Workflows
OpenDiscoveryTrace is a public dataset of 558 complete AI scientific agent trajectories that records the reasoning process—thoughts, tool calls, observations, errors, revision triggers, and confidence—across 124 scientific tasks in drug discovery, materials science, genomics, and literature analysis. The dataset includes seven models (three frontier models and four open‑weight models) and 60 live‑retrieval variants, providing a balanced view of performance and error patterns. Pilot analysis shows that process traces reveal behavioral differences invisible to output‑only evaluation, such as differing error rates and types among frontier models.
arXiv AI
Aug 25
K-Bench: measuring model performance on real scientific agent requests
arXiv:2608.21601v1 Announce Type: new Abstract: Benchmarks for scientific artificial intelligence are mostly written to be scored: multiple-choice questions, curated agent tasks with reference soluti...