A token-level, training-free steering framework that improves LLM personalized co-writing by mixing the base model's posterior with a kernel-density estimate from a small user dataset.
Selective Prompt Anchoring for Code Generation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Recent advances in large language models (LLMs) have transformed software development by automatically generating code from natural language. Yet challenges remain in generating fully correct code that aligns with user intent. Our study reveals that LLMs tend to pay less attention to user prompts as more code tokens are generated. We hypothesize that this attention dilution issue is an important reason for code generation errors. To mitigate this issue, we propose Selective Prompt Anchoring (SPA) to guide code LLMs to pay more attention to user intent when generating code. We evaluate SPA using six base LLMs across six benchmarks. Our results demonstrate that SPA enhances Pass@1 by up to 12.9%, consistently outperforming SOTA code generation methods in all settings. Our code is available at https://github.com/magic-YuanTian/Selective-Prompt-Anchoring.
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Training-Free Token-Level Steering for LLM Personalized Co-Writing
A token-level, training-free steering framework that improves LLM personalized co-writing by mixing the base model's posterior with a kernel-density estimate from a small user dataset.