38 citations · 148 across the 39 of their papers we have counts for
45 papers
World of SWECraft: Agent Collaboration at Scale via Simulated Software Worlds
Shannon Zejiang Shen, Yapei Chang, Qiyao Liang +13
As teams of AI agents take on larger projects and long-standing open problems, agent collaboration offers another dimension for scaling inference-time compute. To understand what m…
Context Language Models
Rulin Shao, Shannon Zejiang Shen, Junjie Oscar Yin +10
We introduce Context Language Models (CLMs), language models that natively manage their own context. We implement this by treating the context as a file and allowing the model to m…
When Agents Slow Down: Understanding LLM Agents' Test-Time Strategies via Elo-per-token Analysis
Kaiyuan Liu, Qiuyang Mang, Bo Peng +6
Large language model (LLM) agents allocate test-time compute adaptively as they revise solutions, use tools, explore alternatives, and decide when to stop. This test-time strategy…
Data Scarcity and Model Sparsity: Mixtures-of-Experts Overfit More to Repeated Data
Atindra Jha, Margaret Li, Jure Leskovec +2
As the supply of human-written text is exhausted, it has become standard practice to repeat language model training data. Prior work has studied data repetition for densely activat…
Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall
Jacqueline He, Howard Yen, Shuyue Stella Li +9
Logit-based knowledge distillation (KD) is used to train smaller language models (LMs) via supervision from stronger teachers, but whether its benefits are consistent across traini…
SPADE: Self-Play in Adaptive Synthetic Executable Environments
Bo Liu, Simon Yu, Yiding Jiang +15
Continuous self-improvement requires an ever-expanding pool of self-generated, diverse, adaptive goals. For language agents, existing training environment pools (hand-curated, stat…
Prefix Sliding for efficient test-time scaling
Niklas Muennighoff, Zhengyang Wang, Zeyi Chen +15
Test-time scaling uses extra test-time compute to improve performance, such as letting language models reason longer when solving a problem. As models keep the entire reasoning tra…
Spokes: Optimizing for Diverse Pretraining Data Selection
Clarence Lee, Yejin Choi, Luke Zettlemoyer +2
Diversity plays a critical role in data selection, improving performance under fixed data budgets by reducing redundancy and repetition. However, optimizing for diversity is inhere…
Scaling Participation in Modular AI Systems
Shangbin Feng, Yike Wang, Weijia Shi +3
Humanity is a mosaic of multifaceted talents and needs, and any truly intelligent AI must reflect that richness. Yet the LLMs used by all are built by the few -- a centralized mark…
JobBench: Aligning Agent Work With Human Will
Yuetai Li, Yichen Feng, Zhangchen Xu +21
Current benchmarks for occupational AI agents are scoped primarily by economic values, telling a replacement story. We introduce JobBench, which evaluates AI agents on the workflow…
Slicing and Dicing: Configuring Optimal Mixtures of Experts
Margaret Li, Sneha Kudugunta, Danielle Rothermel +1
Mixture-of-Experts (MoE) architectures have become standard in large language models, yet many of their core design choices - expert count, granularity, shared experts, load balanc…
Micro Language Models Enable Instant Responses
Wen Cheng, Tuochao Chen, Karim Helwani +3
Edge devices such as smartwatches and smart glasses cannot continuously run even the smallest 100M-1B parameter language models due to power and compute constraints, yet cloud infe…
Gecko: An Efficient Neural Architecture Inherently Processing Sequences with Arbitrary Lengths
Xuezhe Ma, Shicheng Wen, Linghao Jin +11
Designing a unified neural network to efficiently and inherently process sequential data with arbitrary lengths is a central and challenging problem in sequence modeling. The desig…
MoCo: A One-Stop Shop for Model Collaboration Research
Shangbin Feng, Yuyang Bai, Ziyuan Yang +17
Advancing beyond single monolithic language models (LMs), recent research increasingly recognizes the importance of model collaboration, where multiple LMs collaborate, compose, an…
Self-Improving VLM Judges Without Human Annotations
Inna Wanyin Lin, Yushi Hu, Shuyue Stella Li +5
Effective judges of Vision-Language Models (VLMs) are crucial for model development. Current methods for training VLM judges mainly rely on large-scale human preference annotations…
Separate the Wheat from the Chaff: Winnowing Down Divergent Views in Retrieval Augmented Generation
Song Wang, Zihan Chen, Peng Wang +5
Retrieval-augmented generation (RAG) enhances large language models (LLMs) by integrating external knowledge sources to address their limitations in accessing up-to-date or special…
Train for Truth, Keep the Skills: Binary Retrieval-Augmented Reward Mitigates Hallucinations
Tong Chen, Akari Asai, Luke Zettlemoyer +2
Language models often generate factually incorrect information unsupported by their training data, a phenomenon known as extrinsic hallucination. Existing mitigation approaches oft…
FlexOlmo: Open Language Models for Flexible Data Use
Weijia Shi, Akshita Bhagia, Kevin Farhat +20
We introduce FlexOlmo, a new class of language models (LMs) that supports (1) distributed training without data sharing, where different model parameters are independently trained…
Spurious Rewards: Rethinking Training Signals in RLVR
Rulin Shao, Shuyue Stella Li, Rui Xin +11
We show that reinforcement learning with verifiable rewards (RLVR) can elicit strong mathematical reasoning in certain language models even with spurious rewards that have little,…
Precise Information Control in Long-Form Text Generation
Jacqueline He, Howard Yen, Margaret Li +7
A central challenge in language models (LMs) is faithfulness hallucination: the generation of information unsubstantiated by input context. To study this problem, we propose Precis…
ReasonIR: Training Retrievers for Reasoning Tasks
Rulin Shao, Rui Qiao, Varsha Kishore +8
We present ReasonIR-8B, the first retriever specifically trained for general reasoning tasks. Existing retrievers have shown limited gains on reasoning tasks, in part because exist…
ParaPO: Aligning Language Models to Reduce Verbatim Reproduction of Pre-training Data
Tong Chen, Faeze Brahman, Jiacheng Liu +5
Language models (LMs) can memorize and reproduce segments from their pretraining data verbatim even in non-adversarial settings, raising concerns about copyright, plagiarism, priva…
s1: Simple test-time scaling
Niklas Muennighoff, Zitong Yang, Weijia Shi +7
Test-time scaling is a promising new approach to language modeling that uses extra test-time compute to improve performance. Recently, OpenAI's o1 model showed this capability but…
Heterogeneous Swarms: Jointly Optimizing Model Roles and Weights for Multi-LLM Systems
Shangbin Feng, Zifeng Wang, Palash Goyal +8
We propose Heterogeneous Swarms, an algorithm to design multi-LLM systems by jointly optimizing model roles and weights. We represent multi-LLM systems as directed acyclic graphs (…