전체 글419 [2026-1] 이루가 - Lawformer: A Pre-trained Language Model for Chinese Legal Long Documents 논문 링크: https://arxiv.org/abs/2105.03887 Lawformer: A Pre-trained Language Model for Chinese Legal Long DocumentsLegal artificial intelligence (LegalAI) aims to benefit legal systems with the technology of artificial intelligence, especially natural language processing (NLP). Recently, inspired by the success of pre-trained language models (PLMs) in the generic domaiarxiv.org ABSTRACTLawformer는 기.. 2026. 3. 21. [2026-1] 김지원 - SPICE: Self-Play In Corpus Environments 논문 제목: SPICE: Self-Play In Corpus Environments논문 링크: https://arxiv.org/abs/2510.24684 SPICE: Self-Play In Corpus Environments Improves ReasoningSelf-improving systems require environmental interaction for continuous adaptation. We introduce SPICE (Self-Play In Corpus Environments), a reinforcement learning framework where a single model acts in two roles: a Challenger that mines documents from a.. 2026. 3. 21. [2026-1] 김학선, 임준수 - BENCHMARKING AGENTS ON HARD, REALISTICTASKS IN COMMAND LINE INTERFACES https://arxiv.org/abs/2601.11868 Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line InterfacesAI agents may soon become capable of autonomously completing valuable, long-horizon tasks in diverse domains. Current benchmarks either do not measure real-world tasks, or are not sufficiently difficult to meaningfully measure frontier models. To this end,arxiv.org Task Formula.. 2026. 3. 20. [2026-1] 정유림 - Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference paper : https://arxiv.org/abs/2412.13663hugging face : https://huggingface.co/docs/transformers/main/en/model_doc/modernbert기존 BERT 모델의 한계를 뛰어넘는 ModernBERT를 소개합니다. 2조 개의 토큰으로 학습된 이 모델은 8192 토큰의 긴 시퀀스 길이에서도 뛰어난 성능을 발휘하며, 분류 및 검색 작업에서 최첨단 결과를 달성합니다. 또한, 가장 빠르고 메모리 효율적인 인코더로서 일반적인 GPU에서 효율적인 추론이 가능하도록 설계되었습니다.1. ModernBERT: 기존 BERT의 한계를 뛰어넘는 혁신적인 인코더 모델ModernBERT는 2조 개의 토큰으로 학습되어 8192 토큰의 긴 시퀀스 길이에.. 2026. 3. 20. 이전 1 2 3 4 5 6 7 8 ··· 105 다음
