Energy-Based Reasoners:循環式推理模型應採用能量模型
Energy-Based Reasoners: Your Recurrent Reasoner Should Be an Energy-Based Model
Energy-Based Reasoners(EBRs)以能量函數及其梯度引導循環式潛在推理狀態更新,並提供停止訊號。在四項算法推理基準中,EBRs 的表現媲美或優於先前的循環式推理模型,在 Sudoku-Extreme 達到 99.9% 準確率,所用推理計算量比次佳模型少 345 倍。在所有多解基準中,EBRs 均達到最先進的不同有效解覆蓋率,同時維持接近 100% 的單次樣本準確率。
Published on Nov 6
Authors:
,
,
,
Abstract
Energy-Based Reasoners guide looped latent reasoning by learning an energy function whose gradient refines states and signals when to stop, with exploration-based training improving performance and efficiency.
Looping is a powerful inductive bias for reasoning. By repeatedly iterating a recurrent block, looped models can scale their effective depth and enable multi-step reasoning in latent space. However, existing looped models are not inherently guided by an explicit measure of progress, making it difficult to determine when to stop looping and how to improve upon the current state. To enable this, we introduce Energy-Based Reasoners (EBRs), which learn an energy function over latent reasoning states and formulate looped updates as gradient descent on this energy. The energy scores latent states, while its gradient guides refinement and provides a signal for stopping. To train EBRs effectively, we further introduce exploration-based training to discover additional modes and to better shape the energy landscape over a broader set of states. Across four algorithmic reasoning benchmarks, EBRs match or outperform prior recurrent reasoning models, achieving 99.9% accuracy on Sudoku-Extreme while using 345times less inference compute than the next-best model and maintaining comparable training compute. On all multi-solution benchmarks, EBRs achieve state-of-the-art coverage of distinct valid solutions while maintaining near-100% single-sample accuracy. These results suggest that energy-based models are a promising approach for strong reasoning performance.
View arXiv page View PDF Project page GitHub Add to collection
Get this paper in your agent:
hf papers read 2611.08977
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2611.08977 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2611.08977 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2611.08977 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co