跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 3 天前AI 評分51

在大語言模型中置入 CPU:以零梯度編譯將精確電路寫入預訓練權重

A CPU Inside an LLM: Zero-Gradient Compilation of Exact Circuits into Pretrained Weights

AI 導讀

研究提出零梯度電路嫁接方法,無需訓練樣本或執行時掛鈎,便可把確定性計算直接編入預訓練語言模型權重。在 6 位數算術測試(N=1,200,BF16)中,IBM Granite 4.2 3B 和 Alibaba Qwen 3.5 4B 的 ALU-prefix 準確率分別為 99.17% 和 100.00%。

正文

Published on Nov 19

Authors:

Abstract

Pretrained language models (LMs) remain unreliable on exact multi-digit arithmetic without gradient fine-tuning or external tool interpreters. We introduce closed-form, zero-gradient circuit-grafting primitives that compile deterministic computation directly into pretrained LM weights using zero training examples and zero runtime hooks, while bounding per-layer normalization drift and preserving natural-language (NL) competence. Our construction combines three analytical mechanisms: (1) two-layer Mellin-GLU quadrature over bias-free SwiGLU/GeGLU layers, turning unknown per-token RMSNorm scaling into translation along a log-spaced gate grid (>250times dynamic range) with IEEE 754 error bounds; (2) cross-token scale cancellation via signed Chebyshev Indefinite-RoPE self-sink attention and memoryless SiLU-antisymmetric GatedDeltaNet shifts; and (3) axis-aligned width extension (d to d+K) with d/(d+K) norm compensation, Base-1000 canonical BPE emission, and differential (+254,-256) tied unembeddings. An automated 5-stage compiler synthesizes 4-8 digit signed {+,-,times} circuits in constant TC^0 depth alongside 3-literal Boolean (100%, N=1{,}728; plus 42-layer Gemma 4 E4B audits) and 5-bit Majority/Parity (100%) circuits. On 6-digit arithmetic (N=1{,}200, BF16), IBM Granite 4.2 3B (1.07M-1.33M active non-zeros, le 0.036%) and Alibaba Qwen 3.5 4B (2.25M-2.39M non-zeros, le 0.057%) reach 99.17% and 100.00% ALU-prefix accuracy (100% with an FP32 sidecar) and 98%-100% mid-sentence resumption, outperforming 12.8K-sample LoRA (42.2% / 61.7%, 0% inline continuation). A factorized block-sparse sidecar holds A100 memory at 7.89 / 10.65 GiB (4.60 / 8.14 GiB under INT8/INT4 + FP32 sidecar) while passing all 14/14 pre-registered >2K-token Tango splits (N=4{,}392; 0/11,134 OOD triggers).

View arXiv page View PDF Add to collection

Get this paper in your agent:

hf papers read 2611.26450

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/2611.26450 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/2611.26450 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/2611.26450 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