JW
Jason Weston
cs.CLcs.AIcs.LGArtificial IntelligenceAlgorithmsstat.MLcs.IRcs.CVcs.HCcs.MA
On Valency
published · living versionsW_h5g9ucen·v1 · currentpublished
The Goldilocks Principle: Reading Children's Books with Explicit Memory Representations
with Felix Hill, Antoine Bordes, Sumit Chopra
1 version
Preprints & journals
162 papers in the corpus · 2002–2025Training Large Language Models to Reason in a Continuous Latent Space2412.06769v4 · Shibo Hao, Sainbayar Sukhbaatar, DiJia Su et al.2024 · 17 citationsarXiv
The Alignment Waltz: Jointly Training Agents to Collaborate for Safety2510.08240v2 · Jingyu Zhang, Haozhu Wang, Eric Michael Smith et al.2025 · 0 citationsarXiv
Learning to Solve and Verify: A Self-Play Framework for Code and Test Generation2502.14948v4 · Zi Lin, Sheng Shen, Ilia Kulikov et al.2025 · 0 citationsarXiv
AI & Human Co-Improvement for Safer Co-Superintelligence2512.05356v2 · Jason Weston, Jakob Foerster2025 · 0 citationsarXiv
Scaling Agent Learning via Experience Synthesis2511.03773v2 · Zhaorun Chen, Zhuokai Zhao, Kai Zhang et al.2025 · 0 citationsarXiv
NaturalReasoning: Reasoning in the Wild with 2.8M Challenging Questions2502.13124v4 · Weizhe Yuan, Jane Yu, Song Jiang et al.2025 · 0 citationsarXiv
Hybrid Reinforcement: When Reward Is Sparse, It's Better to Be Dense2510.07242v3 · Leitian Tao, Ilia Kulikov, Swarnadeep Saha et al.2025 · 0 citationsarXiv
J1: Incentivizing Thinking in LLM-as-a-Judge via Reinforcement Learning2505.10320v3 · Chenxi Whitehouse, Tianlu Wang, Ping Yu et al.2025 · 1 citationarXiv
OptimalThinkingBench: Evaluating Over and Underthinking in LLMs2508.13141v2 · Pranjal Aggarwal, Seungone Kim, Jack Lanchantin et al.2025 · 1 citationarXiv
The Majority is not always right: RL training for solution aggregation2509.06870v1 · Wenting Zhao, Pranjal Aggarwal, Swarnadeep Saha et al.2025 · 0 citationsarXiv
CoT-Self-Instruct: Building high-quality synthetic prompts for reasoning and non-reasoning tasks2507.23751v2 · Ping Yu, Jack Lanchantin, Tianlu Wang et al.2025 · 0 citationsarXiv
Jointly Reinforcing Diversity and Quality in Language Model Generations2509.02534v1 · Tianjian Li, Yiming Zhang, Ping Yu et al.2025 · 1 citationarXiv
Source2Synth: Synthetic Data Generation and Curation Grounded in Real Data Sources2409.08239v2 · Alisia Lupidi, Carlos Gemmell, Nicola Cancedda et al.2024 · 0 citationsarXiv
Meta CLIP 2: A Worldwide Scaling Recipe2507.22062v3 · Yung-Sung Chuang, Yang Li, Dong Wang et al.2025 · 5 citationsarXiv
Learning to Plan & Reason for Evaluation with Thinking-LLM-as-a-Judge2501.18099v2 · Swarnadeep Saha, Xian Li, Marjan Ghazvininejad et al.2025 · 0 citationsarXiv
Self-Consistency Preference Optimization2411.04109v3 · Archiki Prasad, Weizhe Yuan, Richard Yuanzhe Pang et al.2024 · 0 citationsarXiv
NaturalThoughts: Selecting and Distilling Reasoning Traces for General Reasoning Tasks2507.01921v1 · Yang Li, Youssef Emad, Karthik Padthe et al.2025 · 0 citationsarXiv
Bridging Offline and Online Reinforcement Learning for LLMs2506.21495v1 · Jack Lanchantin, Angelica Chen, Janice Lan et al.2025 · 0 citationsarXiv
Self-Challenging Language Model Agents2506.01716v1 · Yifei Zhou, Sergey Levine, Jason Weston et al.2025 · 1 citationarXiv
Diverse Preference Optimization2501.18101v4 · Jack Lanchantin, Angelica Chen, Shehzaad Dhuliawala et al.2025 · 1 citationarXiv
Self-Rewarding Language Models2401.10020v3 · Weizhe Yuan, Richard Yuanzhe Pang, Kyunghyun Cho et al.2024 · 10 citationsarXiv
SWEET-RL: Training Multi-Turn LLM Agents on Collaborative Reasoning Tasks2503.15478v1 · Yifei Zhou, Song Jiang, Yuandong Tian et al.2025 · 0 citationsarXiv
R.I.P.: Better Models by Survival of the Fittest Prompts2501.18578v2 · Ping Yu, Weizhe Yuan, Olga Golovneva et al.2025 · 0 citationsarXiv
An Overview of Large Language Models for Statisticians2502.17814v1 · Wenlong Ji, Weizhe Yuan, Emily Getzen et al.2025 · 9 citationsarXiv
LLM Pretraining with Continuous Concepts2502.08524v1 · Jihoon Tack, Jack Lanchantin, Jane Yu et al.2025 · 0 citationsarXiv
Step-KTO: Optimizing Mathematical Reasoning through Stepwise Binary Feedback2501.10799v1 · Yen-Ting Lin, Di Jin, Tengyu Xu et al.2025 · 1 citationarXiv
Byte Latent Transformer: Patches Scale Better Than Tokens2412.09871v1 · Artidoro Pagnoni, Ram Pasunuru, Pedro Rodriguez et al.2024 · 19 citationsarXiv
ALMA: Alignment with Minimal Annotation2412.04305v1 · Michihiro Yasunaga, Leonid Shamis, Chunting Zhou et al.2024 · 0 citationsarXiv
Adaptive Decoding via Latent Preference Optimization2411.09661v1 · Shehzaad Dhuliawala, Ilia Kulikov, Ping Yu et al.2024 · 1 citationarXiv
Thinking LLMs: General Instruction Following with Thought Generation2410.10630v1 · Tianhao Wu, Janice Lan, Weizhe Yuan et al.2024 · 3 citationsarXiv
Backtracking Improves Generation Safety2409.14586v1 · Yiming Zhang, Jianfeng Chi, Hailey Nguyen et al.2024 · 0 citationsarXiv
Better Alignment with Instruction Back-and-Forth Translation2408.04614v2 · Thao Nguyen, Jeffrey Li, Sewoong Oh et al.2024 · 2 citationsarXiv
Self-Taught Evaluators2408.02666v2 · Tianlu Wang, Ilia Kulikov, Olga Golovneva et al.2024 · 3 citationsarXiv
Meta-Rewarding Language Models: Self-Improving Alignment with LLM-as-a-Meta-Judge2407.19594v2 · Tianhao Wu, Weizhe Yuan, Olga Golovneva et al.2024 · 11 citationsarXiv
Distilling System 2 into System 12407.06023v3 · Ping Yu, Jing Xu, Jason Weston et al.2024 · 1 citationarXiv
Iterative Reasoning Preference Optimization2404.19733v3 · Richard Yuanzhe Pang, Weizhe Yuan, Kyunghyun Cho et al.2024 · 4 citationsarXiv
Following Length Constraints in Instructions2406.17744v1 · Weizhe Yuan, Ilia Kulikov, Ping Yu et al.2024 · 3 citationsarXiv
Branch-Solve-Merge Improves Large Language Model Evaluation and Generation2310.15123v2 · Swarnadeep Saha, Omer Levy, Asli Celikyilmaz et al.2023 · 11 citationsarXiv
Contextual Position Encoding: Learning to Count What's Important2405.18719v2 · Olga Golovneva, Tianlu Wang, Jason Weston et al.2024 · 3 citationsarXiv
Reverse Training to Nurse the Reversal Curse2403.13799v3 · Olga Golovneva, Zeyuan Allen-Zhu, Jason Weston et al.2024 · 1 citationarXiv
Some things are more CRINGE than others: Iterative Preference Optimization with the Pairwise Cringe Loss2312.16682v2 · Jing Xu, Andrew Lee, Sainbayar Sukhbaatar et al.2023 · 0 citationsarXiv
TOOLVERIFIER: Generalization to New Tools via Self-Verification2402.14158v2 · Dheeraj Mekala, Jason Weston, Jack Lanchantin et al.2024 · 3 citationsarXiv
Self-Alignment with Instruction Backtranslation2308.06259v3 · Xian Li, Ping Yu, Chunting Zhou et al.2023 · 13 citationsarXiv
Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM2403.07816v1 · Sainbayar Sukhbaatar, Olga Golovneva, Vasu Sharma et al.2024 · 5 citationsarXiv
System-Level Natural Language Feedback2306.13588v3 · Weizhe Yuan, Kyunghyun Cho, Jason Weston2023 · 1 citationarXiv
Leveraging Implicit Feedback from Deployment Data in Dialogue2307.14117v2 · Richard Yuanzhe Pang, Stephen Roller, Kyunghyun Cho et al.2023 · 1 citationarXiv
System 2 Attention (is something you might need too)2311.11829v1 · Jason Weston, Sainbayar Sukhbaatar2023 · 12 citationsarXiv
The ART of LLM Refinement: Ask, Refine, and Trust2311.07961v1 · Kumar Shridhar, Koustuv Sinha, Andrew Cohen et al.2023 · 7 citationsarXiv
Learning to Reason and Memorize with Self-Notes2305.00833v2 · Jack Lanchantin, Shubham Toshniwal, Jason Weston et al.2023 · 5 citationsarXiv
Walking Down the Memory Maze: Beyond Context Limit through Interactive Reading2310.05029v1 · Howard Chen, Ramakanth Pasunuru, Jason Weston et al.2023 · 9 citationsarXiv
Career total: 326 works. 162 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.