Research map / Large language models / Agents, code and math
Graph of States: Solving Abductive Tasks with Large Language Models
Yu Luo, Rongchen Gao, Lu Teng, Xidao Wen, Jiamin Jiang, Qingliang Zhang, Yongqian Sun, Shenglin Zhang, Jiasong Feng, Tong Liu, Wenjie Zhang, Dan Pei
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add reasoning focus (depth-first on top hypothesis) (ablation) | GoS · DiagnosisArena medical diagnosis | Match (%): 19.57 -> 31.88 | Improved |
| add explicit causal graph (ablation) | GoS · DiagnosisArena medical diagnosis | Match (%): 12.32 -> 31.88 | Improved |
| add state machine (ablation) | GoS · DiagnosisArena medical diagnosis | Match (%): 12.32 -> 31.88 | Improved |
| add state machine (ablation) | GoS · DiagnosisArena medical diagnosis | Cost ($/case): 0.17 -> 0.12 | Improved |
| use causal graph instead of structured-text state management (ablation) | GoS · DiagnosisArena medical diagnosis | Match (%): 18.12 -> 31.88 | Improved |
| structured text state management over no causal graph (ablation) | GoS · DiagnosisArena medical diagnosis | Match (%): 12.32 -> 18.12 | Improved |
| use GoS instead of Multi/FoT | GoS · Failure diagnosis, Lenovo | Match (%): 28.00 -> 70.67 | Improved |
| use GoS instead of Multi/FoT | GoS · DiagnosisArena medical diagnosis | Cost ($/case): 0.73 -> 0.12 | Improved |
| use multi-agent instead of single-agent baseline | CoT · DiagnosisArena medical diagnosis | Relevant (%): 47.83 -> 49.28 | Improved |
| use ToT instead of CoT | single-agent baseline · DiagnosisArena medical diagnosis | Match (%): 21.01 -> 18.84 | Hurt |
| extend topology from tree to forest (FoT) | multi-agent baseline · DiagnosisArena medical diagnosis | Relevant (%): 50.72 -> 63.04 | Improved |
| raise drill-down thresholds eta and delta too high (ablation) | GoS · DiagnosisArena medical diagnosis | Match vs Relevant | Mixed |
Closest papers by meaning
- Boule or Baguette? A Study on Task Topology, Length Generalization, and the Benefit of Reasoning Trace (NeurIPS 2026 · similarity 0.87)
- Topology of Reasoning: Understanding Large Reasoning Models through Reasoning Graph Properties (NeurIPS 2025 · similarity 0.87)
- Analysis for Abductive Learning and Neural-Symbolic Reasoning Shortcuts (ICML 2024 · similarity 0.87)
- NeSTR: A Neuro-Symbolic Abductive Framework for Temporal Reasoning in Large Language Models (AAAI 2026 · similarity 0.87)
- Reasoning by Superposition: A Theoretical Perspective on Chain of Continuous Thought (NeurIPS 2025 · similarity 0.87)
- System-1.5 Reasoning: Traversal in Language and Latent Spaces with Dynamic Shortcuts (NeurIPS 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
