paper-with-me

Papers

Easy Problems That LLMs Get Wrong

2024-05-30 · Sean Williams, James Huckle

We introduce a comprehensive Linguistic Benchmark designed to evaluate the limitations of Large Language Models (LLMs) in domains such as logical reasoning, spatial intelligence, and linguistic understanding, among others. Through a series of straightforward questions, it uncovers the significant limitations of well-regarded models to perform tasks that humans manage with ease. It also highlights the potential of prompt engineering to mitigate some errors and underscores the necessity for better training methodologies. Our findings stress the importance of grounding LLMs with human reasoning and common sense, emphasising the need for human-in-the-loop for enterprise applications. We hope this work paves the way for future research to enhance the usefulness and reliability of new models.

📄 PDF Abstract BibTeX arXiv:2405.19616

Code (1)

autogenai/easy-problems-that-llms-get-wrong 공식 구현

Tasks

Common Sense ReasoningLogical ReasoningPrompt Engineering

Similar Papers 제목 키워드 기반

Varying Shades of Wrong: Aligning LLMs with Wrong Answers Only

2024-10-14 · Jihan Yao, Wenxuan Ding, Shangbin Feng, Lucy Lu Wang 외

In the absence of abundant reliable annotations for challenging tasks and contexts, how can we expand the frontier of LLM capabilities with potentially wrong answers? We focus on two research questions: (1) Can LLMs gene…

PhD Knowledge Not Required: A Reasoning Challenge for Large Language Models

2025-02-03 · Zixuan Wu, Francesca Lucchetti, Aleksander Boruch-Gruszecki, Jingmiao Zhao 외

Existing benchmarks for frontier models often test specialized, "PhD-level" knowledge that is difficult for non-experts to grasp. In contrast, we present a benchmark with 594 problems based on the NPR Sunday Puzzle Chall…

General Knowledge

Learning to Solve Hard Problems in RL for LLMs by Never Giving Up

2026-09-11 · Michael Noukhovitch, Hamish Ivison, Nathan Lambert, Aaron Courville hf

We demonstrate that training LLMs with RL does not improve performance equally across a dataset. RL shows large improvements on easy problems that an LLM is already good at solving, but small improvements on hard problem…

KCSAT-ML: Probing Reasoning Models with Nationwide-Cohort Human Difficulty

2026-06-09 · Sanghee Park, Geewook Kim, Kee-Eung Kim arxiv

Math reasoning benchmarks have proliferated, yet most lack a per-item difficulty signal grounded in actual human performance. We introduce KCSAT-ML, a decade (2014-2025) of Korean College Scholastic Ability Test (KCSAT; …

Wrong-of-Thought: An Integrated Reasoning Framework with Multi-Perspective Verification and Wrong Information

2024-10-06 · Yongheng Zhang, Qiguang Chen, Jingxuan Zhou, Peng Wang 외

Chain-of-Thought (CoT) has become a vital technique for enhancing the performance of Large Language Models (LLMs), attracting increasing attention from researchers. One stream of approaches focuses on the iterative enhan…