Towards Context-Robust LLMs: A Gated Representation Fine-tuning Approach
This addresses the issue of misleading and unhelpful contexts in LLMs for applications like RAG, though it appears incremental as it builds on existing fine-tuning methods.
The paper tackles the problem of LLMs over-relying on imperfect external contexts in retrieval-augmented generation, proposing a gated representation fine-tuning approach that achieves context-robust behaviors by training a lightweight intervention function with only 0.0004% of model size on fewer than 200 examples.
Large Language Models (LLMs) enhanced with external contexts, such as through retrieval-augmented generation (RAG), often face challenges in handling imperfect evidence. They tend to over-rely on external knowledge, making them vulnerable to misleading and unhelpful contexts. To address this, we propose the concept of context-robust LLMs, which can effectively balance internal knowledge with external context, similar to human cognitive processes. Specifically, context-robust LLMs should rely on external context only when lacking internal knowledge, identify contradictions between internal and external knowledge, and disregard unhelpful contexts. To achieve this goal, we introduce Grft, a lightweight and plug-and-play gated representation fine-tuning approach. Grft consists of two key components: a gating mechanism to detect and filter problematic inputs, and low-rank representation adapters to adjust hidden representations. By training a lightweight intervention function with only 0.0004\% of model size on fewer than 200 examples, Grft can effectively adapt LLMs towards context-robust behaviors.