Related papers: SPICE: Self-Play In Corpus Environments Improves R…
Existing dense retrieval models struggle with reasoning-intensive retrieval task as they fail to capture implicit relevance that requires reasoning beyond surface-level semantic information. To address these challenges, we propose…
The dominant paradigm for improving mathematical reasoning in language models relies on Reinforcement Learning with verifiable rewards. Yet existing methods treat each problem instance in isolation without leveraging the reusable strategies…
Many real-world tasks require language models (LMs) to reason over complex contexts that exceed their parametric knowledge. This calls for context learning, where LMs directly learn relevant knowledge from the given context. An intuitive…
In scenarios where language models must incorporate new information efficiently without extensive retraining, traditional fine-tuning methods are prone to overfitting, degraded generalization, and unnatural language generation. To address…
Reinforcement learning (RL) has proven effective for fine-tuning large language models (LLMs), significantly enhancing their reasoning abilities in domains such as mathematics and code generation. A crucial factor influencing RL fine-tuning…
An important challenge in reinforcement learning is training agents that can solve a wide variety of tasks. If tasks depend on each other (e.g. needing to learn to walk before learning to run), curriculum learning can speed up learning by…
Recent advancements in long-context large language models have attracted significant attention, yet their practical applications often suffer from suboptimal context utilization. This study investigates structuring training data to enhance…
Text corpus exploration (TCE) spans the range of exploratory search tasks: it goes beyond simple retrieval to include item discovery and learning about the corpus and topic. Systems support TCE with tools such as similarity-based…
Reinforcement learning has advanced video reasoning in large multi-modal models, yet dominant pipelines either rely on on-policy self-exploration, which plateaus at the model's knowledge boundary, or hybrid replay that mixes policies and…
Deploying language-model agents in production often requires substantial compute and human effort to tune prompts, parsers, validators, and other components of the agent pipeline. Self-evolution offers a promising alternative, but most…
Robot learning has proven to be a general and effective technique for programming manipulators. Imitation learning is able to teach robots solely from human demonstrations but is bottlenecked by the capabilities of the demonstrations.…
Prompt-based models have demonstrated impressive prompt-following capability at image editing tasks. However, the models still struggle with following detailed editing prompts or performing local edits. Specifically, global image quality…
Large language models are able to learn new tasks in context, where they are provided with instructions and a few annotated examples. However, the effectiveness of in-context learning is dependent on the provided context, and the…
This work investigates the reasoning and planning capabilities of foundation models and their scalability in complex, dynamic environments. We introduce PuzzlePlex, a benchmark designed to assess these capabilities through a diverse set of…
Reinforcement learning is a widely used approach to autonomous navigation, showing potential in various tasks and robotic setups. Still, it often struggles to reach distant goals when safety constraints are imposed (e.g., the wheeled robot…
High-quality labeled datasets are crucial for training and evaluating foundation models in software engineering, but creating them is often prohibitively expensive and labor-intensive. We introduce SPICE, a scalable, automated pipeline for…
Understanding and reasoning about physics is an important ability of intelligent agents. We develop the PHYRE benchmark for physical reasoning that contains a set of simple classical mechanics puzzles in a 2D physical environment. The…
Self-play with large language models has emerged as a promising paradigm for achieving self-improving artificial intelligence. However, existing self-play frameworks often suffer from optimization instability, due to (i) non-stationary…
This work-in-progress paper presents SPARC (Systematic Problem Solving and Algorithmic Reasoning for Children), a gamified learning platform designed to enhance engagement and knowledge retention in K-12 STEM education. Traditional…
Throughout long history, natural species have learned to survive by evolving their physical structures adaptive to the environment changes. In contrast, current reinforcement learning (RL) studies mainly focus on training an agent with a…