MKA: Leveraging Cross-Lingual Consensus for Model Abstention
Not on the map. Why?
Abstract (arXiv)
Reliability of LLMs is questionable even as they get better at more tasks. A wider adoption of LLMs is contingent on whether they are usably factual. And if they are not, on whether they can properly calibrate their confidence in their responses. This work focuses on utilizing the multilingual knowledge of an LLM to inform its decision to abstain or answer when prompted. We develop a multilingual pipeline to calibrate the model's confidence and let it abstain when uncertain. We run several multilingual models through the pipeline to profile them across different languages. We find that the performance of the pipeline varies by model and language, but that in general they benefit from it. This is evidenced by the accuracy improvement of 71.2% for Bengali over a baseline performance without the pipeline. Even a high-resource language like English sees a 15.5% improvement. These results hint at possible further improvements.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 18 of its 20 references that are in the map. First posted on arXiv 31 Mar 2025.
Builds on
18 of 20 references.
References · 18
- 2020Measuring Massive Multitask Language Understandingcited by 7,963
- 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,998
- 2022Language Models (Mostly) Know What They Knowcited by 1,856
- 2022No Language Left Behind: Scaling Human-Centered Machine Translationcited by 1,312
- 2023SGLang: Efficient Execution of Structured Language Model Programscited by 1,141
- 2022When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memoriescited by 1,128
- 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 974
- 2020Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillationcited by 593
- 2020How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answeringcited by 445
- 2024Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researcherscited by 363
- 2024Don't Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaborationcited by 180
- 2024Fine-grained Hallucination Detection and Editing for Language Modelscited by 149
6 more
- 2024Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontiercited by 129
- 2024Know Your Limits: A Survey of Abstention in Large Language Modelscited by 121
- 2023Do Multilingual Language Models Think Better in English?cited by 88
- 2023Towards Mitigating Hallucination in Large Language Models via Self-Reflectioncited by 81
- 2024Reliable, Adaptable, and Attributable Language Models with Retrievalcited by 73
- 2024Teaching LLMs to Abstain across Languages via Multilingual Feedbackcited by 14
Built on by
Citing papers per year: 2025: 2.
2 papers.