paper-with-me

Papers

EpiK-Eval: Evaluation for Language Models as Epistemic Models

2023-10-23 · Gabriele Prato, Jerry Huang, Prasannna Parthasarathi, Shagun Sodhani, Sarath Chandar

In the age of artificial intelligence, the role of large language models (LLMs) is becoming increasingly central. Despite their growing prevalence, their capacity to consolidate knowledge from different training documents - a crucial ability in numerous applications - remains unexplored. This paper presents the first study examining the capability of LLMs to effectively combine such information within their parameter space. We introduce EpiK-Eval, a novel question-answering benchmark tailored to evaluate LLMs' proficiency in formulating a coherent and consistent knowledge representation from segmented narratives. Evaluations across various LLMs reveal significant weaknesses in this domain. We contend that these shortcomings stem from the intrinsic nature of prevailing training objectives. Consequently, we advocate for refining the approach towards knowledge consolidation, as it harbors the potential to dramatically improve their overall effectiveness and performance. The findings from this study offer insights for developing more robust and reliable LLMs. Our code and benchmark are available at https://github.com/chandar-lab/EpiK-Eval

📄 PDF Abstract BibTeX arXiv:2310.15372

Code (1)

chandar-lab/epik-eval 공식 구현 pytorch

Tasks

Question Answering

Similar Papers 제목 키워드 기반

Everyday Physics in Korean Contexts: A Culturally Grounded Physical Reasoning Benchmark

2025-09-22 · Jihae Jeong, DaeYeop Lee, DongGeon Lee, Hwanjo Yu arxiv

Existing physical commonsense reasoning benchmarks predominantly focus on Western contexts, overlooking cultural variations in physical problem-solving. To address this gap, we introduce EPiK (Everyday Physics in Korean …

Physical Commonsense Reasoning

EPIK: Eliminating multi-model Pipelines with Knowledge-distillation

2022-11-27 · Bhavesh Laddagiri, Yash Raj, Anshuman Dash

Real-world tasks are largely composed of multiple models, each performing a sub-task in a larger chain of tasks, i.e., using the output from a model as input for another model in a multi-model pipeline. A model like MATR…

Knowledge DistillationTransliteration

Leveraging System-Level Observations to Inform Bayesian Learning of Model Parameters for Quantitative Verification

2026-08-04 · Simos Gerasimou, Xingyu Zhao arxiv

Combining Bayesian learning and quantitative verification is a powerful toolset for analysing key quantitative properties of software systems, like reliability and response time. However, the accuracy and robustness of v…

Bayesian Inference

ClinicalBench: Stress-Testing Assertion-Aware Retrieval for Cross-Admission Clinical QA on MIMIC-IV

2026-05-11 · Alex Stinard arxiv

Reasoning benchmarks measure clinical performance on clean inputs. We evaluate the step before reasoning: retrieval over real EHR notes, where negation, temporality, and family-versus-patient attribution can flip a corre…

Are LLM-Judges Robust to Expressions of Uncertainty? Investigating the effect of Epistemic Markers on LLM-based Evaluation

2024-10-28 · Dongryeol Lee, Yerin Hwang, Yongil Kim, Joonsuk Park 외

In line with the principle of honesty, there has been a growing effort to train large language models (LLMs) to generate outputs containing epistemic markers. However, evaluation in the presence of epistemic markers has …