相关论文: Scaling Laws for Agent Harnesses via Effective Fee…

As scientific applications extend to the simulation of more and more complex systems, they involve an increasing number of abstraction levels, at each of which errors can emerge and across which they can propagate; tools for correctness…

Multi-agent systems (MAS) are increasingly capable of tackling complex real-world tasks, yet their reliance on inter-agent coordination, tool use, and long-horizon reasoning makes error recognition particularly challenging. Minor errors can…

Effective automation hinges on deciding when to act and when to escalate. We model this as a decision under uncertainty: an LLM forms a prediction, estimates its probability of being correct, and compares the expected costs of acting and…

We explore the impact of parameter sparsity on the scaling behavior of Transformers trained on massive datasets (i.e., "foundation models"), in both vision and language domains. In this setting, we identify the first scaling law describing…

Agent communication remains a foundational problem in multi-agent systems: protocols such as FIPA-ACL guarantee semantic richness but are intractable for constrained environments, while lightweight IoT protocols achieve efficiency at the…

Large language models have achieved strong performance on medical reasoning benchmarks, yet their deployment in clinical settings demands rigorous verification to ensure factual accuracy. While reward models offer a scalable approach for…

While large language models have significantly accelerated scientific code generation, comprehensively evaluating the generated code remains a major challenge. Traditional benchmarks reduce evaluation to test-case matching, an approach…

Recent advances in Transformer-based large language models (LLMs) have led to significant performance improvements across many tasks. These gains come with a drastic increase in the models' size, potentially leading to slow and costly use…

Agent evaluation requires assessing complex multi-step behaviors involving tool use and intermediate reasoning, making it costly and expertise-intensive. A natural question arises: can frontier coding assistants reliably automate this…

While Large Language Models (LLMs) have substantially improved the functional correctness of code translation, the critical dimension of \textit{execution efficiency} remains overlooked. We present \textbf{\textsc{trace}}, the first…

While Large Language Models (LLMs) have substantially improved the functional correctness of code translation, the critical dimension of \textit{execution efficiency} remains overlooked. We present \textbf{\textsc{trace}}, the first…

Large Language Models (LLMs) are increasingly acting as autonomous agents, with function calling (FC) capabilities enabling them to invoke specific tools for tasks. While prior research has primarily focused on improving FC accuracy, little…

Model Predictive Control (MPC) can efficiently control constrained systems in real-time applications. MPC feedback law for a linear system with linear inequality constraints can be explicitly computed off-line, which results in an off-line…

This work addresses the synthesis of optimal feedback control laws via machine learning. In particular, the Averaged Feedback Learning Scheme (AFLS) and a data driven method are considered. Hypotheses for each method ensuring the…

To address the complexities posed by time- and state-varying uncertainties and the computation of analytic derivatives in strict-feedback form (SFF) systems, this study introduces a novel model reference-based control (MRBC) framework which…

As AI agents surpass human capabilities, scalable oversight -- the problem of effectively supplying human feedback to potentially superhuman AI models -- becomes increasingly critical to ensure alignment. While numerous scalable oversight…

Reward Models (RMs) are crucial for aligning language models with human preferences. Currently, the evaluation of RMs depends on measuring accuracy against a validation set of manually annotated preference data. Although this method is…

Reliable effort estimation remains an ongoing challenge to software engineers. Accurate effort estimation is the state of art of software engineering, effort estimation of software is the preliminary phase between the client and the…

Scaling up neural models has yielded significant advancements in a wide array of tasks, particularly in language generation. Previous studies have found that the performance of neural models frequently adheres to predictable scaling laws,…

Inference-time compute scaling has emerged as a powerful technique for improving the reliability of large language model (LLM) agents, but existing methods apply compute uniformly: every decision step receives the same budget regardless of…