跳到正文
arXiv cs.AI· Allen G. Roush, Minh Nhat Nguyen, Ravid Shwartz-Ziv, Judah Goldfeder, Sanjay Basu·· 4 小时前AI 评分58

论文称长上下文生成退化是采样问题,top-nsigma在高温下保持准确率

Long-Context Generation Is a Sampling Problem (When You Turn Up the Heat)

AI 导读

论文指出开源权重模型生成越长越差的问题在高温度下主要由采样器决定,而非需要更多训练。作者测试了29个1B到122B参数的开源模型,覆盖SWE-bench Verified编码智能体。

正文 · 原文

View PDF HTML (experimental)

Abstract:Open-weight language models are widely observed to get worse the longer they generate. Often, attempts to fix this involve more training. At high temperature, The sampler, i.e. the rule that picks the next token out of the model's probabilities, decides whether long and agentic generation survives and/or thrives. We test 29 open-weight models from 1B to 122B parameters on ten established benchmarks plus needle retrieval out to one million tokens, covering SWE-bench Verified coding agents, long chain-of-thought reasoning, long-form writing, long structured output and long-context reading. Samplers that keep their pool of candidate tokens small and steady as temperature rises don't lose accuracy. Samplers that let in low-probability tokens fall apart. At temperature 2, coding agents running top-p, top-k or Top-H resolve 0 of 100 GitHub issues, while the same agents running top-nsigma resolve 46 to 62 and win every issue the two samplers disagree on, on three Qwen3.5 models. Over all benchmarks top-nsigma keeps a median 100% of its temperature-1 score where top-p keeps 26%. Counting how many tokens each sampler lets through explains why. In one Llama-3.1-8B story, going from temperature 1 to 2 grows top-p's pool from 4 tokens to 51,845 while top-nsigma stays at 3. Across 133 combinations of sampler setting and temperature, the typical pool size predicts judged quality (Spearman correlation of -0.77), and when we pick the pool at temperature 1 and only apply temperature afterwards, Top-H, min-p and p-less stop collapsing. At the default temperature of 1 most samplers perform about the same, We release all code and data to the public.
Comments: 10 pages, accepted to the 40th Conference on Neural Information Processing Systems (NeurIPS 2026). Track on Evaluations and Datasets
Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
Cite as: arXiv:2611.22148 [cs.AI]
  (or arXiv:2611.22148v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.22148

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Allen Roush [view email]
[v1] Tue, 17 Nov 2026 03:30:03 UTC (457 KB)

来源:arXiv cs.AI · arxiv.org