Papers with MMLU benchmark
NumeroLogic: Number Encoding for Enhanced LLMs’ Numerical Reasoning (2024.emnlp-main)
Copied to clipboard
| Challenge: | Language models struggle with numerical and arithmetical tasks, such as multiplying 3-digit numbers. |
| Approach: | They propose a method to include the count of digits before each number instead of “42”. |
| Outcome: | The proposed format improves the reasoning process before generating the actual number. |
Investigating Data Contamination in Modern Benchmarks for Large Language Models (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing evaluation benchmarks for large language models are inflated and inconsistent with actual performance. |
| Approach: | They propose a retrieval-based system to explore potential overlaps between benchmarks and pretraining corpora and a protocol to investigate testset slot guessing. |
| Outcome: | The proposed method exploits overlaps between evaluation benchmarks and pretraining corpora and masks a wrong answer in a multiple choice question and prompts the model to fill in the gap. |
SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents (2024.acl-long)
Copied to clipboard
Ruiyi Wang, Haofei Yu, Wenxin Zhang, Zhengyang Qi, Maarten Sap, Yonatan Bisk, Graham Neubig, Hao Zhu
| Challenge: | Existing studies on building language agents have not addressed this social learning gap. |
| Approach: | They propose an interactive learning method that improves the social intelligence of language agents by using behavior cloning and self-reinforcement based training on filtered social interaction data. |
| Outcome: | The proposed method allows a 7B LLM to reach the social goal completion ability of an expert model (GPT-4-based agent) without the loss of more generic abilities, such as the ability to answer knowledge-based questions. |
Calibration Across Layers: Understanding Calibration Evolution in LLMs (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated inherent calibration capabilities, where predicted probabilities align well with correctness . previous studies have linked this behavior to specific components in the final layer, such as entropy neurons and the unembedding matrix’s null space. |
| Approach: | They propose to examine how calibration evolves throughout the network's depth. |
| Outcome: | The proposed calibration direction improves calibration metrics without harming accuracy. |
None of the Above, Less of the Right Parallel Patterns in Human and LLM Performance on Multi-Choice Questions Answering (2025.findings-acl)
Copied to clipboard
| Challenge: | Multiple-choice exam questions with “None of the above” (NA) options have been extensively studied in educational testing . however, their impact on Large Language Models (LLMs) evaluation remains underexplored . |
| Approach: | They conduct systematic experiments with 28 LLMs on the MMLU benchmark to examine how NA options affect model performance and confidence calibration. |
| Outcome: | The results highlight important implications for benchmark design and raise questions about LLMs’ ability to handle uncertainty in real-world applications. |
Forget What You Know about LLMs Evaluations - LLMs are Like a Chameleon (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) excel on public benchmarks, but high scores may mask overreliance on dataset-specific surface cues rather than true language understanding. |
| Approach: | They propose a meta-evaluation framework that systematically rephrases benchmark inputs to detect overfitting. |
| Outcome: | The proposed framework detects performance degradation indicative of superficial pattern reliance on dataset-specific cues and distortion levels. |