When to Speak, When to Abstain: Contrastive Decoding with Abstention
- URL: http://arxiv.org/abs/2412.12527v2
- Date: Sun, 16 Feb 2025 14:09:46 GMT
- Title: When to Speak, When to Abstain: Contrastive Decoding with Abstention
- Authors: Hyuhng Joon Kim, Youna Kim, Sang-goo Lee, Taeuk Kim,
- Abstract summary: Large Language Models (LLMs) demonstrate exceptional performance across diverse tasks by leveraging pre-trained (i.e., parametric) and external (i.e., contextual) knowledge.
We propose Contrastive Decoding with Abstention, a novel training-free decoding method that allows LLMs to generate responses when relevant knowledge is available and to abstain otherwise.
- Score: 12.639085523137998
- License:
- Abstract: Large Language Models (LLMs) demonstrate exceptional performance across diverse tasks by leveraging pre-trained (i.e., parametric) and external (i.e., contextual) knowledge. While substantial efforts have been made to enhance the utilization of both forms of knowledge, situations in which models lack relevant information remain underexplored. To investigate this challenge, we first present a controlled testbed featuring four distinct knowledge access scenarios, including the aforementioned edge case, revealing that conventional LLM usage exhibits insufficient robustness in handling all instances. Addressing this limitation, we propose Contrastive Decoding with Abstention (CDA), a novel training-free decoding method that allows LLMs to generate responses when relevant knowledge is available and to abstain otherwise. CDA estimates the relevance of both knowledge sources for a given input, adaptively deciding which type of information to prioritize and which to exclude. Through extensive experiments, we demonstrate that CDA can effectively perform accurate generation and abstention simultaneously, enhancing reliability and preserving user trust.
Related papers
- IAO Prompting: Making Knowledge Flow Explicit in LLMs through Structured Reasoning Templates [7.839338724237275]
We introduce IAO (Input-Action-Output), a structured template-based method that explicitly models how Large Language Models access and apply their knowledge.
IAO decomposes problems into sequential steps, each clearly identifying the input knowledge being used, the action being performed, and the resulting output.
Our findings provide insights into both knowledge representation within LLMs and methods for more reliable knowledge application.
arXiv Detail & Related papers (2025-02-05T11:14:20Z) - UAlign: Leveraging Uncertainty Estimations for Factuality Alignment on Large Language Models [41.67393607081513]
Large Language Models (LLMs) often struggle to accurately express the factual knowledge they possess.
We propose the UAlign framework, which leverages Uncertainty estimations to represent knowledge boundaries.
We show that the proposed UAlign can significantly enhance the LLMs' capacities to confidently answer known questions.
arXiv Detail & Related papers (2024-12-16T14:14:27Z) - KBAlign: Efficient Self Adaptation on Specific Knowledge Bases [73.34893326181046]
Large language models (LLMs) usually rely on retrieval-augmented generation to exploit knowledge materials in an instant manner.
We propose KBAlign, an approach designed for efficient adaptation to downstream tasks involving knowledge bases.
Our method utilizes iterative training with self-annotated data such as Q&A pairs and revision suggestions, enabling the model to grasp the knowledge content efficiently.
arXiv Detail & Related papers (2024-11-22T08:21:03Z) - Mind the Interference: Retaining Pre-trained Knowledge in Parameter Efficient Continual Learning of Vision-Language Models [79.28821338925947]
Domain-Class Incremental Learning is a realistic but challenging continual learning scenario.
To handle these diverse tasks, pre-trained Vision-Language Models (VLMs) are introduced for their strong generalizability.
This incurs a new problem: the knowledge encoded in the pre-trained VLMs may be disturbed when adapting to new tasks, compromising their inherent zero-shot ability.
Existing methods tackle it by tuning VLMs with knowledge distillation on extra datasets, which demands heavy overhead.
We propose the Distribution-aware Interference-free Knowledge Integration (DIKI) framework, retaining pre-trained knowledge of
arXiv Detail & Related papers (2024-07-07T12:19:37Z) - Large Language Models are Limited in Out-of-Context Knowledge Reasoning [65.72847298578071]
Large Language Models (LLMs) possess extensive knowledge and strong capabilities in performing in-context reasoning.
This paper focuses on a significant aspect of out-of-context reasoning: Out-of-Context Knowledge Reasoning (OCKR), which is to combine multiple knowledge to infer new knowledge.
arXiv Detail & Related papers (2024-06-11T15:58:59Z) - Towards Safer Large Language Models through Machine Unlearning [19.698620794387338]
Selective Knowledge Unlearning ( SKU) is designed to eliminate harmful knowledge while preserving utility on normal prompts.
First stage aims to identify and acquire harmful knowledge within the model, whereas the second is dedicated to remove this knowledge.
Our experiments demonstrate that SKU identifies a good balance point between removing harmful information and preserving utility.
arXiv Detail & Related papers (2024-02-15T16:28:34Z) - Thrust: Adaptively Propels Large Language Models with External Knowledge [58.72867916604562]
Large-scale pre-trained language models (PTLMs) are shown to encode rich knowledge in their model parameters.
The inherent knowledge in PTLMs can be opaque or static, making external knowledge necessary.
We propose the instance-level adaptive propulsion of external knowledge (IAPEK), where we only conduct the retrieval when necessary.
arXiv Detail & Related papers (2023-07-19T20:16:46Z) - Knowledge Rumination for Pre-trained Language Models [77.55888291165462]
We propose a new paradigm dubbed Knowledge Rumination to help the pre-trained language model utilize related latent knowledge without retrieving it from the external corpus.
We apply the proposed knowledge rumination to various language models, including RoBERTa, DeBERTa, and GPT-3.
arXiv Detail & Related papers (2023-05-15T15:47:09Z) - A Unified End-to-End Retriever-Reader Framework for Knowledge-based VQA [67.75989848202343]
This paper presents a unified end-to-end retriever-reader framework towards knowledge-based VQA.
We shed light on the multi-modal implicit knowledge from vision-language pre-training models to mine its potential in knowledge reasoning.
Our scheme is able to not only provide guidance for knowledge retrieval, but also drop these instances potentially error-prone towards question answering.
arXiv Detail & Related papers (2022-06-30T02:35:04Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.