近期论文83
- DeepScientist: Advancing Frontier-Pushing Scientific Findings Progressively第 7 位,共 8 位作者ICLR 2026 · 57 次引用
- AutoFigure: Generating and Refining Publication-Ready Scientific Illustrations第 9 位,共 9 位作者ICLR 2026 · 28 次引用
- Beyond English-Centric Training: How Reinforcement Learning Improves Cross-Lingual Reasoning in LLMs第 4 位,共 4 位作者ICLR 2026 · 11 次引用
- Pre-DPO: Improving Data Utilization in Direct Preference Optimization Using a Guiding Reference Model第 5 位,共 5 位作者AAAI 2026 · 7 次引用
- LAMDAS: LLM as an Implicit Classifier for Domain-specific Data Selection第 7 位,共 7 位作者AAAI 2026 · 1 次引用
- Minimizing Mismatch Risk: A Prototype-Based Routing Framework for Zero-shot LLM-generated Text Detection第 4 位,共 4 位作者ICML 2026 · 1 次引用
- ReEfBench: Quantifying the Reasoning Efficiency of LLMs第 5 位,共 5 位作者ACL 2026 · 1 次引用
- Zero-Shot Text-to-Motion Evaluation using Video Language Models第 3 位,共 3 位作者ICML 2026
- DeepReview: Improving LLM-based Paper Review with Human-like Deep Thinking Process第 4 位,共 4 位作者ACL 2025 · 70 次引用
- Lost in Literalism: How Supervised Training Shapes Translationese in LLMs第 8 位,共 8 位作者ACL 2025 · 12 次引用
- Exploring Model Editing for LLM-based Aspect-Based Sentiment Classification第 4 位,共 5 位作者AAAI 2025 · 9 次引用
- Unveiling Attractor Cycles in Large Language Models: A Dynamical Systems View of Successive Paraphrasing第 5 位,共 5 位作者ACL 2025 · 3 次引用
- ALLabel: Three-stage Active Learning for LLM-based Entity Recognition using Demonstration Retrieval第 5 位,共 5 位作者EMNLP 2025 · 1 次引用
- Learning to Rank for In-Context Example Retrieval第 7 位,共 7 位作者NeurIPS 2025 · 1 次引用
- Multi-Document Event Extraction Using Large and Small Language Models第 6 位,共 6 位作者EMNLP 2025 · 1 次引用
- Reasoning is All You Need for Video Generalization: A Counterfactual Benchmark with Sub-question Evaluation第 8 位,共 8 位作者ACL Findings 2025 · 1 次引用
- Constrain Alignment with Sparse Autoencoders第 10 位,共 11 位作者ICML 2025
- CycleResearcher: Improving Automated Research via Automated Review第 6 位,共 7 位作者ICLR 2025
- ELICIT: LLM Augmentation Via External In-context Capability第 3 位,共 4 位作者ICLR 2025
- Glimpse: Enabling White-Box Methods to Use Proprietary Models for Zero-Shot LLM-Generated Text Detection第 4 位,共 4 位作者ICLR 2025
- NovelQA: Benchmarking Question Answering on Documents Exceeding 200K Tokens第 11 位,共 11 位作者ICLR 2025
- Personality Alignment of Large Language Models第 4 位,共 4 位作者ICLR 2025
- PerSphere: A Comprehensive Framework for Multi-Faceted Perspective Retrieval and Summarization第 8 位,共 8 位作者ACL 2025
- Task Calibration: Calibrating Large Language Models on Inference Tasks第 4 位,共 4 位作者ACL Findings 2025
- Unaligned Message-Passing and Contextualized-Pretraining for Robust Geo-Entity Resolution第 7 位,共 7 位作者AAAI 2025
- PandaLM: An Automatic Evaluation Benchmark for LLM Instruction Tuning Optimization第 14 位,共 14 位作者ICLR 2024 · 368 次引用
- Fast-DetectGPT: Efficient Zero-Shot Detection of Machine-Generated Text via Conditional Probability Curvature第 5 位,共 5 位作者ICLR 2024 · 311 次引用
- AutoSurvey: Large Language Models Can Automatically Write Surveys第 13 位,共 13 位作者NeurIPS 2024 · 151 次引用
- Can Language Models Learn to Skip Steps?第 5 位,共 9 位作者NeurIPS 2024 · 92 次引用
- MAGE: Machine-generated Text Detection in the Wild第 9 位,共 9 位作者ACL 2024 · 44 次引用
展示最近收录的 30 篇论文。
