跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 12 天前AI 評分49

Tree-Ising:以位元級 Ising 編輯修復語言模型

Tree-Ising: Repairing Language Models with Bit-Level Ising Edits

AI 導讀

Tree-Ising 將語言模型權重修復轉化為 BF16 權重上的稀疏 Ising/QUBO 最佳化問題,用於修復推理循環,無需梯度或 LLM tokens。在 Qwen3.5-0.8B 的兩層修復中,Tree-Ising 的 MATH-500 評分為 54.6%,高於原編輯方法的 51.7% 和基礎模型的 7.9%;模擬分岔求解器取得 55.1%。

正文

Published on Nov 10

Authors:

Abstract

Tree-Ising repairs reasoning loops in LLM weights by solving sparse Ising/QUBO problems over BF16 parameters, improving math benchmark accuracy with fewer weight edits and no gradient or LLM-token cost.

Weight-space programming, introduced in recent MIT work by Takehi et al., enables a frontier LLM agent to read a network's weights and write a repair. Their GPT-6 Astra agent corrected the reasoning loops of Qwen3.5-0.8B without gradients, using 158 agent turns, 89.7 million tokens, and 50 forward-pass requests. Our method, Tree-Ising, casts this repair as a sparse discrete optimization problem over the model's own BF16 weights. Forward statistics define an objective that weakens a loop direction, strengthens a concluding direction, and preserves the layer's output on productive tokens. Within a state-tree representation, where lossless BF16 storage acts as a special case, the admissible edits are sign flips and exponent shifts that preserve all shared group exponents and mantissas. We decompose this objective exactly into independent QUBO (Ising) problems, one per output row, which we solve on a GPU using simulated annealing or quantum-inspired simulated bifurcation. On the agent's two layers, Tree-Ising achieves 54.6% on MATH-500, outperforming the agent's edit (51.7%) under the same evaluation protocol, while the base model scores 7.9% as reported. The simulated bifurcation solver yields a statistically indistinguishable repair at 55.1%. Both solvers outperform the agent edit across all five math benchmarks, with statistically significant improvements on MATH-500 and GSM8K after correcting for multiple comparisons. Our approach modifies 13% to 17% fewer weights and reduces BF16 bit flips by a factor of 2.2 to 2.3. It requires only four forward-pass requests and zero LLM tokens, with the annealing recipe costing about $0.90 in GPU time. On Qwen3.5-2B, with layers and dose selected using only the base model, the same recipe increases greedy GSM8K accuracy from 21.6% to 69.4%. As a proof of concept, we execute an eight-variable repair subproblem using QAOA on an IBM Heron processor.

View arXiv page View PDF GitHub 0 Add to collection

Get this paper in your agent:

hf papers read 2611.14210

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/2611.14210 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/2611.14210 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/2611.14210 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