LLMs as Factual Reasoners: Insights from Existing Benchmarks and Beyond

• URL: http://arxiv.org/abs/2305.14540v1
• Date: Tue, 23 May 2023 21:50:06 GMT
• Title: LLMs as Factual Reasoners: Insights from Existing Benchmarks and Beyond
• Authors: Philippe Laban, Wojciech Kry\'sci\'nski, Divyansh Agarwal, Alexander R. Fabbri, Caiming Xiong, Shafiq Joty, Chien-Sheng Wu
• Abstract summary: We propose a new protocol for inconsistency detection benchmark creation and implement it in a 10-domain benchmark called SummEdits. Most LLMs struggle on SummEdits, with performance close to random chance. The best-performing model, GPT-4, is still 8% below estimated human performance.
• Score: 135.8013388183257
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: With the recent appearance of LLMs in practical settings, having methods that can effectively detect factual inconsistencies is crucial to reduce the propagation of misinformation and improve trust in model outputs. When testing on existing factual consistency benchmarks, we find that a few large language models (LLMs) perform competitively on classification benchmarks for factual inconsistency detection compared to traditional non-LLM methods. However, a closer analysis reveals that most LLMs fail on more complex formulations of the task and exposes issues with existing evaluation benchmarks, affecting evaluation precision. To address this, we propose a new protocol for inconsistency detection benchmark creation and implement it in a 10-domain benchmark called SummEdits. This new benchmark is 20 times more cost-effective per sample than previous benchmarks and highly reproducible, as we estimate inter-annotator agreement at about 0.9. Most LLMs struggle on SummEdits, with performance close to random chance. The best-performing model, GPT-4, is still 8\% below estimated human performance, highlighting the gaps in LLMs' ability to reason about facts and detect inconsistencies when they occur.

Related papers

• Beyond the Singular: The Essential Role of Multiple Generations in Effective Benchmark Evaluation and Analysis [10.133537818749291]
  Large language models (LLMs) have demonstrated significant utilities in real-world applications. Benchmark evaluations are crucial for assessing the capabilities of LLMs.
  arXiv  Detail & Related papers  (2025-02-13T03:43:33Z)
• LLM-Powered Benchmark Factory: Reliable, Generic, and Efficient [19.673388630963807]
  We propose an automated and unbiased evaluation framework, structured around four dimensions and ten criteria. Under this framework, we analyze the advantages and weaknesses of directly prompting large language models (LLMs) as generic benchmark generators. We then introduce a series of methods to address the identified weaknesses and integrate them as BenchMaker. Experiments confirm that BenchMaker achieves superior or comparable performance to human-annotated benchmarks on all metrics.
  arXiv  Detail & Related papers  (2025-02-02T06:36:01Z)
• MMLU-CF: A Contamination-free Multi-task Language Understanding Benchmark [57.999567012489706]
  We propose a contamination-free and more challenging benchmark called MMLU-CF. This benchmark reassesses LLMs' understanding of world knowledge by averting both unintentional and malicious data leakage. Our evaluation of mainstream LLMs reveals that the powerful GPT-4o achieves merely a 5-shot score of 73.4% and a 0-shot score of 71.9% on the test set.
  arXiv  Detail & Related papers  (2024-12-19T18:58:04Z)
• Towards Reproducible LLM Evaluation: Quantifying Uncertainty in LLM Benchmark Scores [2.886479348067378]
  We use benchmarks designed for testing large language models' capacity to reason about cardinal directions. We suggest a simple method for cost-effectively quantifying the uncertainty of a benchmark score.
  arXiv  Detail & Related papers  (2024-10-04T15:04:28Z)
• PaCoST: Paired Confidence Significance Testing for Benchmark Contamination Detection in Large Language Models [41.772263447213234]
  Large language models (LLMs) are known to be trained on vast amounts of data, which may unintentionally or intentionally include data from commonly used benchmarks. This inclusion can lead to cheatingly high scores on model leaderboards, yet result in disappointing performance in real-world applications. We introduce PaCoST, a Paired Confidence Significance Testing to effectively detect benchmark contamination in LLMs.
  arXiv  Detail & Related papers  (2024-06-26T13:12:40Z)
• DARG: Dynamic Evaluation of Large Language Models via Adaptive Reasoning Graph [70.79413606968814]
  We introduce Dynamic Evaluation of LLMs via Adaptive Reasoning Graph Evolvement (DARG) to dynamically extend current benchmarks with controlled complexity and diversity. Specifically, we first extract the reasoning graphs of data points in current benchmarks and then perturb the reasoning graphs to generate novel testing data. Such newly generated test samples can have different levels of complexity while maintaining linguistic diversity similar to the original benchmarks.
  arXiv  Detail & Related papers  (2024-06-25T04:27:53Z)
• Inference-Time Decontamination: Reusing Leaked Benchmarks for Large Language Model Evaluation [61.350306618479365]
  Leakage of benchmarks can prevent the accurate assessment of large language models' true performance. We propose Inference-Time Decontamination (ITD) to address this issue. ITD reduces inflated accuracy by 22.9% on GSM8K and 19.0% on MMLU.
  arXiv  Detail & Related papers  (2024-06-20T04:35:59Z)
• MixEval: Deriving Wisdom of the Crowd from LLM Benchmark Mixtures [57.886592207948844]
  We propose MixEval, a new paradigm for establishing efficient, gold-standard evaluation by strategically mixing off-the-shelf benchmarks. It bridges (1) comprehensive and well-distributed real-world user queries and (2) efficient and fairly-graded ground-truth-based benchmarks, by matching queries mined from the web with similar queries from existing benchmarks.
  arXiv  Detail & Related papers  (2024-06-03T05:47:05Z)
• Self-Evaluation Improves Selective Generation in Large Language Models [54.003992911447696]
  We reformulate open-ended generation tasks into token-level prediction tasks. We instruct an LLM to self-evaluate its answers. We benchmark a range of scoring methods based on self-evaluation.
  arXiv  Detail & Related papers  (2023-12-14T19:09:22Z)

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.