Related papers: When Is Enough Not Enough? Illusory Completion in …

The capability to store data about business processes execution in so-called Event Logs has brought to the diffusion of tools for the analysis of process executions and for the assessment of the goodness of a process model. Nonetheless,…

We study the source of uncertainty in DeepSeek R1-32B by analyzing its self-reported verbal confidence on question answering (QA) tasks. In the default answer-then-confidence setting, the model is regularly over-confident, whereas semantic…

Large language models can perform well on many isolated tasks, yet they continue to struggle on multi-turn, long-horizon agentic problems that require skills such as planning, state tracking, and long context processing. In this work, we…

Retriever-augmented instruction-following models are attractive alternatives to fine-tuned approaches for information-seeking tasks such as question answering (QA). By simply prepending retrieved documents in its input along with an…

Any system which performs goal-directed continual learning must not only learn incrementally but process and absorb information incrementally. Such a system also has to understand when its goals have been achieved. In this paper, we…

Large Language Models (LLMs) have recently achieved impressive performance in math and reasoning benchmarks. However, they often struggle with logic problems and puzzles that are relatively easy for humans. To further investigate this, we…

LLMs reliably correct false claims when presented in isolation, yet when the same claims are embedded in task-oriented requests, they often comply rather than correct. We term this failure mode \emph{correction suppression} and construct a…

Instruction following aims to align Large Language Models (LLMs) with human intent by specifying explicit constraints on how tasks should be performed. However, we reveal a counterintuitive phenomenon: instruction following can…

For Large Language Models (LLMs) to be reliably deployed, models must effectively know when not to answer: abstain. Reasoning models, in particular, have gained attention for impressive performance on complex tasks. However, reasoning…

We present evidence of substantial benefit from efficient exploration in gathering human feedback to improve large language models. In our experiments, an agent sequentially generates queries while fitting a reward model to the feedback…

Frontier coding agents solve complex tasks when given complete context but collapse when specifications are incomplete or ambiguous. The bottleneck is not raw capability, but judgment: knowing when to act autonomously and when to ask for…

Tool-using agents often incur substantial computational cost due to long reasoning chains and iterative tool usage. In practical scenarios, many tasks become infeasible under constrained tool environments, where the capabilities required…

Peer prediction mechanisms incentivize agents to truthfully report their signals even in the absence of verification by comparing agents' reports with those of their peers. In the detail-free multi-task setting, agents respond to multiple…

Large language models (LLMs) have recently shown strong progress on scientific reasoning, yet two major bottlenecks remain. First, explicit retrieval fragments reasoning, imposing a hidden "tool tax" of extra tokens and steps. Second,…

We consider a setting for Inverse Reinforcement Learning (IRL) where the learner is extended with the ability to actively select multiple environments, observing an agent's behavior on each environment. We first demonstrate that if the…

Large language models (LLMs) have been widely integrated into information retrieval to advance traditional techniques. However, effectively enabling LLMs to seek accurate knowledge in complex tasks remains a challenge due to the complexity…

Computer-use agents have rapidly improved on real-world tasks such as web navigation, desktop automation, and software interaction, in some cases surpassing human performance. Yet even when the task and model are unchanged, an agent that…

Despite their substantial successes, AI agents continue to face fundamental challenges in terms of trustworthiness. Consider deep research agents, tasked with searching for information relevant to a given topic-while AI agents can perform…

Achieving effective test-time scaling requires models to engage in In-Context Exploration -- the intrinsic ability to generate, verify, and refine multiple reasoning hypotheses within a single continuous context. Grounded in State Coverage…

Real-time voice assistants must revise task state when users interrupt mid-response, but existing spoken-dialog benchmarks largely evaluate turn-based interaction and miss this failure mode. We introduce EchoChain, a controlled benchmark…