43 accepted papers
Unsupervised Reinforcement Learning with Verifiable Rewards (URLVR) offers a pathway for Large Language Models (LLMs) to improve without human supervision. Particularly, many works use model intrinsic information as rewards for URLVR, showing promising improvements, yet their potential and limitatio…
We propose Observer-Actor (ObAct), a novel framework for active vision imitation learning in which the observer moves to optimal visual observations for the actor. We study ObAct on a dual-arm robotic system equipped with wrist-mounted cameras. At test time, ObAct dynamically assigns observer and ac…
Tool-calling is a central component of modern large language model (LLM) agents, equipping them with skills beyond their parametric knowledge. This paper studies tool-calling along two complementary axes: \textbf{effectiveness}, i.e., how this capability is \textit{measured}, and \textbf{efficiency}…
As large language models evolve into tool-augmented agents, a central question remains unresolved: when is external tool use actually justified? Existing agent frameworks typically treat tools as ordinary actions and optimize for task success or reward, offering little principled distinction between…
Reward modeling is essential for aligning large language models with human preferences through reinforcement learning. To provide accurate reward signals, a reward model (RM) should stimulate deep thinking and conduct interpretable reasoning before assigning a score or a judgment. Inspired by recent…
Drug shortages pose critical risks to patient care and healthcare systems worldwide, yet the effectiveness of regulatory interventions remains poorly understood due to information asymmetries in pharmaceutical supply chains. We propose ShortageSim, which addresses this challenge by providing the fir
As large language models (LLMs) demonstrate increasingly advanced capabilities, aligning their behaviors with human values and preferences becomes crucial for their wide adoption. While previous research focuses on general alignment to principles such as helpfulness, harmlessness, and honesty, the n…
In high-stakes domains such as healthcare and finance, effective decision-making demands not just accurate outcomes but transparent and explainable reasoning. However, current language models often lack the structured deliberation needed for such tasks, instead generating decisions and justification
Positional bias in large language models hinders their ability to effectively process long inputs. A prominent example is the “lost in the middle” phenomenon, where LLMs struggle to utilize relevant information situated in the middle of the input. While prior research primarily focuses on single pie…
Leveraging Multi-modal Large Language Models (MLLMs) to create embodied agents offers a promising avenue for tackling real-world tasks. While language-centric embodied agents have garnered substantial attention, MLLM-based embodied agents remain underexplored due to the lack of comprehensive evaluat…
Large Language Models (LLMs) excel in traditional natural language processing tasks but struggle with problems that require complex domain-specific calculations or simulations. While equipping LLMs with external tools to build LLM-based agents can enhance their capabilities, existing approaches lack…
Language model agents excel in long-session planning and reasoning, but existing benchmarks primarily focus on goal-oriented tasks with explicit objectives, neglecting creative adaptation in unfamiliar environments. To address this, we introduce EscapeBench—a benchmark suite of room escape game envi…
Large Language Models (LLMs) have revolutionized Natural Language Processing (NLP) but pose risks of inadvertently exposing copyrighted or proprietary data, especially when such data is used for training but not intended for distribution. Traditional methods address these leaks only after content is
Recent progress in large language models (LLMs) has enabled substantial advances in solving mathematical problems. However, existing benchmarks often fail to reflect real-world complexity, which demand open-ended, interdisciplinary reasoning and integration of computational tools. To address this ga
Large Language Models (LLMs) have shown remarkable capabilities as autonomous agents; yet existing benchmarks either focus on single-agent tasks or are confined to narrow domains, failing to capture the dynamics of multi-agent coordination and competition. In this paper, we introduce MultiAgentBench…
Agents powered by large language models have shown remarkable abilities in solving complex tasks. However, most agent systems remain reactive, limiting their effectiveness in scenarios requiring foresight and autonomous decision-making. In this paper, we tackle the challenge of developing proactive…
Incorporating external context can significantly enhance the response quality of Large Language Models (LLMs). However, real-world contexts often mix relevant information with disproportionate inappropriate content, posing reliability risks. How do LLMs process and prioritize mixed context? To study
Current Large Language Model (LLM) agents demonstrate strong reasoning and tool use capabilities, but often lack self-awareness, failing to balance these approaches effectively. This imbalance leads to **Tool Overuse**, where models unnecessarily rely on external tools for tasks solvable with parame…
Large language models (LLMs) exhibit exceptional capabilities across various tasks but also pose risks by generating harmful content. Existing safety mechanisms, while improving model safety, often lead to overly cautious behavior and fail to fully leverage LLMs’ internal cognitive processes. Inspir
Hallucination is a persistent challenge in large language models (LLMs), where even with rigorous quality control, models often generate distorted facts. This paradox, in which error generation continues despite high-quality training data, calls for a deeper understanding of the underlying LLM mecha…
Understanding alignment techniques begins with comprehending zero-shot generalization brought by instruction tuning, but little of the mechanism has been understood. Existing work has largely been confined to the task level, without considering that tasks are artificially defined and, to LLMs, merel…
Current Large Language Models (LLMs) often undergo supervised fine-tuning (SFT) to acquire tool use capabilities. However, SFT struggles to generalize to unfamiliar or complex tool use scenarios. Recent advancements in reinforcement learning (RL), particularly with R1-like models, have demonstrated…
Ultrasound-guided therapeutic procedures rely heavily on operator skill, leading to variability and high training costs. The shortage of trained ultra-sonographers further exacerbates the issue, increasing workloads and associated health risks. Robotic technology has the potential to effectively tac
In this work, we introduce PianoMime, a framework for training a piano-playing agent using internet demonstrations. The internet is a promising source of large-scale demonstrations for training our robot agents. In particular, for the case of piano-playing, Youtube is full of videos of professional…
Current language model-driven agents often lack mechanisms for effective user participation, which is crucial given the vagueness commonly found in user instructions. Although adept at devising strategies and performing tasks, these agents struggle with seeking clarification and grasping precise use…
Large Language Models (LLMs) have demonstrated remarkable progress in utilizing tools, but their closed-source nature and high inference costs pose limitations on their adaptability, necessitating a valid method that leverages smaller, open-sourced models. In this paper, we introduce Toolink, a comp…
Large Language Models (LLMs) have made significant progress in utilizing tools, but their ability is limited by API availability and the instability of implicit reasoning, particularly when both planning and execution are involved. To overcome these limitations, we propose CREATOR, a novel framework…
Patient enrollment is critical to the success of a clinical trial. In practice, before launching a trial, one of the top priorities is to predict the enrollment rate for different countries, so that one can select clinical sites from the countries with the highest enrollment rates to accelerate pati…
Continual pre-training is the paradigm where pre-trained language models (PLMs) continually acquire fresh knowledge from growing data and gradually get upgraded. Before an upgraded PLM is released, we may have tuned the original PLM for various tasks and stored the adapted weights. However, when tun…
Tensor decompositions are powerful tools for dimensionality reduction and feature interpretation of multidimensional data such as signals. Existing tensor decomposition objectives (e.g., Frobenius norm) are designed for fitting raw data under statistical assumptions, which may not align with downstr…
Out of Scope (OOS) detection in Conversational AI solutions enables a chatbot to handle a conversation gracefully when it is unable to make sense of the end-user query. Accurately tagging a query as out-of-domain is particularly hard in scenarios when the chatbot is not equipped to handle a topic wh…
Recent years have witnessed the prevalent application of pre-trained language models (PLMs) in NLP. From the perspective of parameter space, PLMs provide generic initialization, starting from which high-performance minima could be found. Although plenty of works have studied how to effectively and e…
Low-rank tensor factorization or completion is well-studied and applied in various online settings, such as online tensor factorization (where the temporal mode grows) and online tensor completion (where incomplete slices arrive gradually). However, in many real-world settings, tensors may have more…
Micro-injection is one of the most efficient approaches to deliver foreign materials into zebrafish larva heart. Well-trained micro-injection technicians are required for such operation, and still the efficiency remains low and the work is laborious. As a result, automated micro-injection that is ab
Control barrier functions are widely used to enforce safety properties in robot motion planning and control. However, the problem of constructing barrier functions online and synthesizing safe controllers that can deal with the associated uncertainty has received little attention. This letter invest
Real-world spatio-temporal data is often incomplete or inaccurate due to various data loading delays. For example, a location-disease-time tensor of case counts can have multiple delayed updates of recent temporal slices for some locations or diseases. Recovering such missing or noisy (under-reporte…
Accurate prediction of the transmission of epidemic diseases such as COVID-19 is crucial for implementing effective mitigation measures. In this work, we develop a tensor method to predict the evolution of epidemic trends for many regions simultaneously. We construct a 3-way spatio-temporal tensor (…
Existing tensor factorization methods assume that the input tensor follows some specific distribution (i.e. Poisson, Bernoulli, and Gaussian), and solve the factorization by minimizing some empirical loss functions defined based on the corresponding distribution. However, it suffers from several dra…
Predicting the response of cancer cells to drugs is an important problem in pharmacogenomics. Recent efforts in generation of large scale datasets profiling gene expression and drug sensitivity in cell lines have provided a unique opportunity to study this problem. However, one major challenge is th…
The 3GPP suggests to combine dual polarized (DP) antenna arrays with the double directional (DD) channel model for downlink channel estimation. This combination strikes a good balance between high-capacity communications and parsimonious channel modeling, and also brings limited feedback schemes for…
Parameter estimation of sinusoids is a problem of significance in many practical applications. This problem is revisited through a new alternating direction method of multipliers (ADMM) based approach, where the unknown parameters are estimated from one-bit quantized noisy measurements with time var…
Phase retrieval has recently attracted renewed interest. It is revisited here through a new approach based on nonconvex quadratically constrained quadratic programming (QCQP). A least-squares (LS) formulation is adopted, and a recently developed non-convex QCQP approximation technique called feasibl…
Joint estimation of the directions-of-arrival (DOAs) and frequencies of multiple signals is addressed in this paper. By constructing a set of joint diagonalization matrices, two cost functions that do not require a priori information of the source number are devised for DOA and frequency estimation…