Pith. sign in

REVIEW 12 cited by

Beyond A*: Better Planning with Transformers via Search Dynamics Bootstrapping

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2402.14083 v2 pith:T67M7LZA submitted 2024-02-21 cs.AI

Beyond A*: Better Planning with Transformers via Search Dynamics Bootstrapping

classification cs.AI
keywords searchdynamicsmodeltaskstrainingcomplexplanningsearchformer
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

While Transformers have enabled tremendous progress in various application settings, such architectures still trail behind traditional symbolic planners for solving complex decision making tasks. In this work, we demonstrate how to train Transformers to solve complex planning tasks. This is accomplished by training an encoder-decoder Transformer model to predict the search dynamics of the $A^*$ search algorithm. We fine tune this model to obtain a Searchformer, a Transformer model that optimally solves previously unseen Sokoban puzzles 93.7% of the time, while using up to 26.8% fewer search steps than the $A^*$ implementation that was used for training initially. In our training method, $A^*$'s search dynamics are expressed as a token sequence outlining when task states are added and removed into the search tree during symbolic planning. Searchformer significantly outperforms baselines that predict the optimal plan directly with a 5-10$\times$ smaller model size and a 10$\times$ smaller training dataset. Lastly, we demonstrate how Searchformer scales to larger and more complex decision making tasks with improved percentage of solved tasks and shortened search dynamics.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 12 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Self-Improvement for Fast, High-Quality Plan Generation

    cs.AI 2026-05 unverdicted novelty 7.0

    Self-improvement of a decoder-only transformer yields plans averaging 30% shorter than a source symbolic planner, over 80% optimal where known, with sub-exponential latency scaling.

  2. Exploration and Exploitation Errors Are Measurable for Language Model Agents

    cs.AI 2026-04 unverdicted novelty 7.0

    A policy-agnostic metric and controllable 2D grid environments with task DAGs enable measurement of exploration and exploitation errors in language model agents from observed actions.

  3. Less is More: Recursive Reasoning with Tiny Networks

    cs.LG 2025-10 unverdicted novelty 7.0

    TRM with 7M parameters achieves 45% accuracy on ARC-AGI-1 and 8% on ARC-AGI-2, surpassing most LLMs with under 0.01% of their parameters.

  4. Training Large Language Models to Reason in a Continuous Latent Space

    cs.CL 2024-12 unverdicted novelty 7.0

    Coconut lets LLMs perform reasoning directly in continuous latent space by recycling hidden states as inputs, outperforming standard chain-of-thought on search-intensive logical tasks with better accuracy-efficiency t...

  5. Emergence of Frontier Superposition: M\"obius attractor and Cascade Supervision

    cs.LG 2026-05 unverdicted novelty 6.0

    Gradient descent reaches equal-weight superposition on Erdős-Rényi reachability via a Möbius attractor under S_n symmetry and Cascade Supervision, with a parameter-free decay law matching experimental cosines within 0.02.

  6. Vision-aligned Latent Reasoning for Multi-modal Large Language Model

    cs.CV 2026-02 unverdicted novelty 6.0

    VaLR generates vision-aligned latent tokens before each reasoning step to preserve perceptual cues, improving VSI-Bench accuracy from 33.0% to 52.9%.

  7. Agent Q: Advanced Reasoning and Learning for Autonomous AI Agents

    cs.AI 2024-08 unverdicted novelty 6.0

    Agent Q integrates MCTS-guided search, self-critique, and off-policy DPO to train LLM agents that outperform behavior cloning and reinforced fine-tuning baselines in WebShop and achieve up to 95.4% success in real-wor...

  8. Search, Fail, Recover: A Training Framework for Correction-Aware Reasoning

    cs.AI 2026-07 conditional novelty 5.0

    Pyligent trains LLMs to search, detect failures via task validators, and backtrack to recoverable prefixes, improving solve rates by 13–73 points over gold-only SFT on hidden graphs, Sudoku, and Blocksworld.

  9. Robots Need More than VLA and World Models

    cs.RO 2026-06 unverdicted novelty 5.0

    The paper identifies four missing interfaces (data autolabelling, embodiment retargeting, physics-grounded world models, and video-based reward inference) as the central bottleneck beyond VLA scaling for robot intelligence.

  10. NoisyCoconut: Counterfactual Consensus via Latent Space Reasoning

    cs.LG 2026-05 unverdicted novelty 5.0

    Injecting noise into LLM latent trajectories creates diverse reasoning paths whose agreement acts as a confidence signal for selective abstention, cutting error rates from 40-70% to under 15% on math tasks.

  11. The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes

    cs.CL 2026-06 unverdicted novelty 4.0

    A literature survey that introduces a taxonomy for LLM reasoning paradigms, analyzes methodological trends, and synthesizes failure modes from over 300 papers.

  12. Efficient Test-time Inference for Generative Planning Models with OCL Search

    cs.AI 2026-05 unverdicted novelty 4.0

    Modified OCL search integrates generative rollouts and learned heuristics for efficient inference in planning models across combinatorial domains.