# Re-Search: Dissecting LLM Ideation Through a Two-Player Questioning Game

Source: [arXiv](https://arxiv.org/abs/2611.14487v1)  
Feed7 permalink: https://feed7.dev/p/2611-14487v1-0m0b92n  
Published: 2026-11-10T18:47:22.000Z  
Trust: Needs Review (needs_review)

## Why Included

A study of research-idea discovery finds most model progress happens early, with later questioning often plateauing. For agentic research tools, it suggests candidate generation may be a bigger bottleneck than judging ideas.

## Source Summary

Re-Search evaluates idea discovery as a yes-or-no questioning game built from **200 ICML 2026 papers**. Across six models, nearly all progress happened in the first **30 of 100 questions**, after which most plateaued.

## Practical Implication

For research agents, the results suggest improving candidate generation or bringing in outside ideas may help more than only improving a model’s ability to assess proposals. The study separates ideation from execution, making the search process easier to inspect.

## Agent-Ready Context

Re-Search evaluates idea discovery as a yes-or-no questioning game built from **200 ICML 2026 papers**. Across six models, nearly all progress happened in the first **30 of 100 questions**, after which most plateaued.

For research agents, the results suggest improving candidate generation or bringing in outside ideas may help more than only improving a model’s ability to assess proposals. The study separates ideation from execution, making the search process easier to inspect.

This is a preprint evaluation of rediscovering published papers’ core ideas; it does not show that the same pattern holds for open-ended research work or deployed coding agents.

## Connected Context

Feed7 judgment across 1059 accumulated Signals:

The early plateau makes the supply of candidate ideas a separate bottleneck from judging them: extending the same questioning process may add little after initial progress. This sharpens where ideation agents might need intervention, while the paper’s rediscovery setup does not establish the pattern for open-ended agent work.

- [Direct Pass@$k$ Optimization for Thought Pruning: Balancing Quality and Diversity in Test-Time Reasoning](https://feed7.dev/p/2610-31667v1-1jz3l6f) — Both focus on the candidate pool: Re-Search finds progress plateaus during sequential idea discovery, while Direct Pass@$k optimizes which diverse reasoning branches remain available. The latter concerns continuations, not idea generation.
- [Legibility is Not Interpretability: Comparing Judged and Actual Importance in Chain-Of-Thought Reasoning](https://feed7.dev/p/2609-04194v1-0milybu) — Re-Search separates proposing ideas from assessing them; the critique study cautions that readable assessments do not reliably identify which reasoning steps matter. Together they discourage treating evaluation ability as a substitute for stronger idea generation.

## Context Map

- Layer: benchmark
- Domains: research
- Topics: agent-evals, reasoning

## Uncertainty

- This is a preprint evaluation of rediscovering published papers’ core ideas; it does not show that the same pattern holds for open-ended research work or deployed coding agents.

## Agent Instruction

Use this item as source-backed context. Do not invent claims beyond the linked source. If this item conflicts with another source, call out the conflict.
