Challenge: Large language models (LLMs) have impressive capabilities, but still suffer from inconsistency issues.
Approach: They develop a ConsisEval benchmark to evaluate LLMs' inconsistency . they find that LLM models can paradoxically fail at easier problems .
Outcome: The proposed model achieves highest consistency score but inconsistent to specific questions due to distraction by redundant information, misinterpretation of questions, etc.

Similar Papers

SCORE: Systematic COnsistency and Robustness Evaluation for Large Language Models (2025.naacl-industry)

Copied to clipboard

Challenge: Typical evaluations of Large Language Models (LLMs) report a single accuracy metric per dataset, often derived from an optimized setup.
Approach: They propose a framework for non-adversarial evaluation of large language models that evaluates models by repeatedly testing them on the same benchmarks in various setups.
Outcome: The proposed framework evaluates models by repeatedly testing them on the same benchmarks in various setups to give a realistic estimate of their accuracy and consistency.
Examining Inter-Consistency of Large Language Models Collaboration: An In-depth Analysis via Debate (2023.findings-emnlp)

Copied to clipboard

Challenge: Existing studies focus on inconsistency issues within a single LLM, while we explore the inter-consistencies among multiple LLMs for collaboration.
Approach: They propose a formal debate framework to examine whether LLMs can collaborate effectively to achieve a consensus for a shared goal.
Outcome: The proposed framework enables LLMs to achieve consensus in three real-world debate scenarios with real-time scenarios aligned to the LLM's goals.
A Systematic Survey and Critical Review on Evaluating Large Language Models: Challenges, Limitations, and Recommendations (2024.emnlp-main)

Copied to clipboard

Challenge: Large Language Models (LLMs) have gained significant attention due to their capabilities in performing diverse tasks across domains.
Approach: They review the primary challenges and limitations causing inconsistencies in evaluations . early models could generate coherent text but limited to simple tasks .
Outcome: The proposed evaluations are reproducible, reliable, and robust.
Are Large Language Models Consistent over Value-laden Questions? (2024.findings-emnlp)

Copied to clipboard

Challenge: Large language models (LLMs) appear to bias survey answers toward certain values . however, some argue that LLMs are inconsistent to simulate particular values - a recent study .
Approach: They define value consistency as similarity of answers across paraphrases, related questions and multilingual translations of a question to English, Chinese, German, and Japanese.
Outcome: The proposed model is consistent across paraphrases, use-cases, translations, and within a topic.
Have LLMs Advanced Enough? A Challenging Problem Solving Benchmark For Large Language Models (2023.emnlp-main)

Copied to clipboard

Challenge: The performance of large language models (LLMs) on existing reasoning benchmarks has significantly improved over the past decade.
Approach: They propose a benchmark dataset for evaluating the problem solving abilities of large language models (LLMs) they curate 515 challenging problems from the highly competitive IIT JEE-Advanced exam.
Outcome: The proposed model performs better on open-source and proprietary models than the current model, but with techniques like self-consistency, self-refinement and chain-of-thought prompting.
SummEdits: Measuring LLM Ability at Factual Reasoning Through The Lens of Summarization (2023.emnlp-main)

Copied to clipboard

Challenge: Existing factual consistency benchmarks are inadequate to detect factual inconsistencies in LLMs.
Approach: They propose a protocol for inconsistency detection benchmark creation and implement it in a 10-domain benchmark called SummEdits.
Outcome: The proposed method is 20 times more cost-effective per sample and highly reproducible, as it estimates inter-annotator agreement at about 0.9.
Are Your LLMs Capable of Stable Reasoning? (2025.findings-acl)

Copied to clipboard

Challenge: Existing evaluation protocols and metrics do not capture the full spectrum of LLM capabilities, especially in complex reasoning tasks.
Approach: They propose a new evaluation metric that continuously assesses model performance across multiple sampling attempts, quantifying both the model’s potential capabilities and operational consistency.
Outcome: The proposed evaluation metric measures model performance across multiple sampling attempts and provides comprehensive insights into their potential capabilities and operational consistency.
Rating Roulette: Self-Inconsistency in LLM-As-A-Judge Frameworks (2025.findings-emnlp)

Copied to clipboard

Challenge: Using large language models (LLMs) for evaluating natural language generation has gained traction . lm judges have low intra-rater reliability in their assigned scores, making it difficult to measure how good their judgments actually are.
Approach: They show that large language models align more closely with human preferences than n-grams . they quantify this variance and compare them to other NLG tasks and benchmarks based on the results .
Outcome: The proposed models align more closely with human preferences than n-gram or embedding-based metrics.
There’s No Such Thing as Simple Reasoning for LLMs (2025.findings-acl)

Copied to clipboard

Challenge: Existing work has focused on relatively complex “many-hop” reasoning problems.
Approach: They analyse the performance of fine-tuned LLMs on simple reasoning problems . they find the models remain highly brittle, being susceptible to seemingly innocent perturbations .
Outcome: The proposed models fail on simple reasoning problems, but are highly brittle . they are susceptible to seemingly innocent perturbations, such as adding duplicates to the set of premises and shuffling the order in which the premises are presented.
A Survey of Confidence Estimation and Calibration in Large Language Models (2024.naacl-long)

Copied to clipboard

Challenge: Large language models (LLMs) have demonstrated impressive capabilities across a wide range of tasks in various domains, but they can be unreliable due to factual errors in their generations.
Approach: They summarize recent advances in LLM confidence estimation and calibration and outline their main lessons learned.
Outcome: The proposed methods can be used to assess the reliability of models and to calibrate them across tasks.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations