TIGER: Learning from Researcher Taste to Generate Tailored Research Ideas
cs.AI
Submitted: 2026-11-01
Updated: 2026-11-01
Project page: https://colah.github.io/notes/taste
Terminology
Sources
- Ideation Arena: Evaluating LLM Generated Research Ideas with Battle-style Human Expert Assessment
- Measuring the Gap Between Human and LLM Research Ideas
- LLMs learn scientific taste from institutional traces across the social sciences
- GIANTS: Generative Insight Anticipation from Scientific Literature
- Writing-Zero: Bridge the Gap Between Non-verifiable Tasks and Verifiable Rewards
- A Measure of Research Taste
- HypoBench: Towards Systematic and Principled Benchmarking for Hypothesis Generation
- The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery
- EvoIdeator: Evolving Scientific Ideas through Checklist-Grounded Reinforcement Learning
- Rethinking Rubric Generation for Improving LLM Judge and Reward Modeling for Open-ended Tasks
- AI Can Learn Scientific Taste
- LigBench: A Unified and Human-Aligned Benchmark for LLM-based Research Idea Generation
- The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search
- ResearchTown: Simulator of Human Research Community
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents