# Negative Knowledge as Failure-aware Shared Memory for AutoResearch

arXiv: https://arxiv.org/abs/2606.21024 · HTML: https://mlsanity.com/paper/2606.21024/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 23 references that are in the map.
- First posted on arXiv 19 Jun 2026.

## Builds on (most cited first, 8 of 8)

- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [Teaching Large Language Models to Self-Debug](https://mlsanity.com/paper/2304.05128/) (2023)
- [ChemCrow: Augmenting large-language models with chemistry tools](https://mlsanity.com/paper/2304.05376/) (2023)
- [Safe Reinforcement Learning via Shielding](https://mlsanity.com/paper/1708.08611/) (2017)
- [Robust Probabilistic Model Checking with Continuous Reward Domains](https://mlsanity.com/paper/2502.04530/) (2025)

## Abstract (arXiv)

AI-assisted research systems generate many failed attempts, but those failures rarely become a durable, shared knowledge asset. We propose a negative knowledge memory layer: a curator agent converts each failed attempt into a bounded, typed record in a shared bank, and a downstream research agent explicitly adopts or rejects those records before proposing its next experiment. We evaluate this layer in two settings: same-task retry on ScienceAgentBench and cross-task scientific research on two nonlinear math-physics PDE problems. The negative knowledge layer outperforms vanilla AutoResearch baselines while using fewer tokens; agents with the negative knowledge bank solve new tasks that all baselines fail to solve in PDE systems research. We also show that the previous negative knowledge bank can transfer and enhance AutoResearch on different PDE problems. These results suggest that structured negative knowledge is a knowledge asset that should be explicitly maintained in broader AI-engaged scientific research beyond a memory-compression or debugging aid, alongside positive findings, as a collective infrastructure for scientific memory. Code is available at https://github.com/hch-wang/Negative_Knowledge.