Truth-Aware Context Selection: Mitigating Hallucinations of Large Language Models Being Misled by Untruthful Contexts (2403.07556v4)
Abstract: Although LLMs have demonstrated impressive text generation capabilities, they are easily misled by untruthful contexts provided by users or knowledge augmentation tools, leading to hallucinations. To alleviate LLMs from being misled by untruthful context and take advantage of knowledge augmentation, we propose Truth-Aware Context Selection (TACS), a lightweight method to adaptively recognize and mask untruthful context from the inputs. TACS begins by performing truth detection on the input context, leveraging the parameterized knowledge within the LLM. Subsequently, it constructs a corresponding attention mask based on the truthfulness of each position, selecting the truthful context and discarding the untruthful context. Additionally, we introduce a new evaluation metric, Disturbance Adaption Rate, to further study the LLMs' ability to accept truthful information and resist untruthful information. Experimental results indicate that TACS can effectively filter untruthful context and significantly improve the overall quality of LLMs' responses when presented with misleading information.
- Evaluating correctness and faithfulness of instruction-following models for question answering.
- Self-consuming generative models go mad.
- A multitask, multilingual, multimodal evaluation of chatgpt on reasoning, hallucination, and interactivity.
- Language models are few-shot learners. ArXiv, abs/2005.14165.
- Language models are few-shot learners.
- Dola: Decoding by contrasting layers improves factuality in large language models. arXiv preprint arXiv:2309.03883.
- On the origin of hallucinations in conversational models: Is it the datasets or the models? In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 5271–5285, Seattle, United States. Association for Computational Linguistics.
- Bias and fairness in large language models: A survey.
- Did aristotle use a laptop? a question answering benchmark with implicit reasoning strategies.
- Critic: Large language models can self-correct with tool-interactive critiquing.
- Sillm: Large language models for simultaneous machine translation.
- Support vector machines. IEEE Intelligent Systems and their Applications, 13(4):18–28.
- Survey of hallucination in natural language generation. ACM Computing Surveys, 55(12):1–38.
- Internet-augmented language models through few-shot prompting for open-domain question answering.
- Factuality enhanced language models for open-ended text generation.
- Inference-time intervention: Eliciting truthful answers from a language model.
- Truthfulqa: Measuring how models mimic human falsehoods.
- Eran Malach. 2023. Auto-regressive next-token predictors are universal learners.
- When not to trust language models: Investigating effectiveness of parametric and non-parametric memories.
- Sources of hallucination by large language models on inference tasks.
- Augmented language models: a survey.
- Factscore: Fine-grained atomic evaluation of factual precision in long form text generation.
- Webgpt: Browser-assisted question-answering with human feedback.
- OpenAI. 2023. Gpt-4 technical report.
- Med-halt: Medical domain hallucination test for large language models.
- In-context retrieval-augmented language models.
- Investigating the factual knowledge boundary of large language models with retrieval augmentation.
- John Schulman. 2023. Reinforcement learning from human feedback: Progress and challenges. Technical report.
- Trusting your evidence: Hallucinate less with context-aware decoding.
- Moss: Training conversational language models from synthetic data.
- Llama 2: Open foundation and fine-tuned chat models.
- A stitch in time saves nine: Detecting and mitigating hallucinations of llms by validating low-confidence generation.
- Attention is all you need.
- Adaptive chameleon or stubborn sloth: Revealing the behavior of large language models in knowledge conflicts. In Proceedings of ICLR.
- Llm lies: Hallucinations are not bugs, but features as adversarial examples.
- Bayling: Bridging cross-lingual alignment and instruction following through interactive translation for large language models.
- Truthx: Alleviating hallucinations by editing large language models in truthful space.
- Siren’s song in the ai ocean: A survey on hallucination in large language models.
- Judging llm-as-a-judge with mt-bench and chatbot arena.
- Lima: Less is more for alignment.
- Representation engineering: A top-down approach to ai transparency.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.