paper-with-me

홈 › Papers

HardSATGEN: Understanding the Difficulty of Hard SAT Formula Generation and A Strong Structure-Hardness-Aware Baseline

2023-02-04 · Yang Li, Xinyan Chen, Wenxuan Guo, Xijun Li, Wanqian Luo, Junhua Huang, Hui-Ling Zhen, Mingxuan Yuan, Junchi Yan

Industrial SAT formula generation is a critical yet challenging task. Existing SAT generation approaches can hardly simultaneously capture the global structural properties and maintain plausible computational hardness. We first present an in-depth analysis for the limitation of previous learning methods in reproducing the computational hardness of original instances, which may stem from the inherent homogeneity in their adopted split-merge procedure. On top of the observations that industrial formulae exhibit clear community structure and oversplit substructures lead to the difficulty in semantic formation of logical structures, we propose HardSATGEN, which introduces a fine-grained control mechanism to the neural split-merge paradigm for SAT formula generation to better recover the structural and computational properties of the industrial benchmarks. Experiments including evaluations on private and practical corporate testbed show the superiority of HardSATGEN being the only method to successfully augment formulae maintaining similar computational hardness and capturing the global structural properties simultaneously. Compared to the best previous methods, the average performance gains achieve 38.5% in structural statistics, 88.4% in computational metrics, and over 140.7% in the effectiveness of guiding solver tuning by our generated instances. Source code is available at http://github.com/Thinklab-SJTU/HardSATGEN

📄 PDF Abstract BibTeX arXiv:2302.02104

Code (1)

thinklab-sjtu/hardsatgen 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Harder Is Better: Boosting Mathematical Reasoning via Difficulty-Aware GRPO and Multi-Aspect Question Reformulation

2026-01-28 · Yanqi Dai, Yuxiang Ji, Xiao Zhang, Yong Wang 외 arxiv

Reinforcement Learning with Verifiable Rewards (RLVR) offers a robust mechanism for enhancing mathematical reasoning in large models. However, we identify a systematic lack of emphasis on more challenging questions in ex…

Reinforcement LearningMathematical Reasoning

TaskEval: Assessing Difficulty of Code Generation Tasks for Large Language Models

2024-07-30 · Florian Tambon, Amin Nikanjam, Cyrine Zid, Foutse khomh 외

Large Language Models (LLMs) excel in code-related tasks like code generation, but benchmark evaluations often overlook task characteristics, such as difficulty. Moreover, benchmarks are usually built using tasks describ…

BenchmarkingCode CompletionCode GenerationHumanEval

Difficulty Translation in Histopathology Images

2020-04-27 · Jerry Wei, Arief Suriawinata, Xiaoying Liu, Bing Ren 외

The unique nature of histopathology images opens the door to domain-specific formulations of image translation models. We propose a difficulty translation model that modifies colorectal histopathology images to be more c…

BIG-bench Machine LearningTranslation

Your Pretrained Model Tells the Difficulty Itself: A Self-Adaptive Curriculum Learning Paradigm for Natural Language Understanding

2025-07-13 · Qi Feng, Yihong Liu, Hinrich Schütze arxiv

Curriculum learning is a widely adopted training strategy in natural language processing (NLP), where models are exposed to examples organized by increasing difficulty to enhance learning efficiency and performance. Howe…

Natural Language UnderstandingMulti-class Classification

Three Questions Concerning the Use of Large Language Models to Facilitate Mathematics Learning

2023-10-20 · An-Zi Yen, Wei-Ling Hsu

Due to the remarkable language understanding and generation abilities of large language models (LLMs), their use in educational applications has been explored. However, little work has been done on investigating the peda…

Mathematical Problem-SolvingPosition