Archive, newest first. Filter by topic, impact and company →
Problem The paper addresses a significant gap in the capabilities of medical agents, which are typically constrained by their pre-deployment design. This limitation restricts their adaptability and effectiveness in dynamic…
{'Problem': 'The paper addresses the significant uncertainty in agent trajectories during reasoning and acting processes, which is critical for improving the reliability of AI systems. This work is particularly relevant…
Problem The paper addresses a significant gap in the capability of existing simulation infrastructures for embodied AI, specifically the need for scalable simulation that facilitates robot data generation, policy training,…
Problem The paper addresses a gap in the capability of low-cost mobile solutions for the early detection and monitoring of various health conditions. It emphasizes the need for accessible diagnostic…
Problem This work addresses the vulnerability of large language models to adversarial manipulation through prompt injection and jailbreak attacks. The authors highlight a gap in understanding how these attacks can…
Problem This work addresses a gap in the preservation of answer-supporting rationales during post-training quantization (PTQ) for medical large language models (LLMs). The authors highlight that while existing quantization methods…
Problem The paper addresses the AI population governance problem, which involves determining how various AI instantiations can be grouped and how their configurations can be effectively represented and monitored. This…
Multiverse Computing has introduced a novel approach to pruning large language models (LLMs) by framing the block removal process as an Ising optimization problem. This method was applied to the…
Recent findings in AI safety research highlight the significant role of language in shaping the harmfulness of AI outputs, asserting that language has a 2.5 times greater impact on safety…
Google Deepmind has introduced a novel method called Dream-RSI, aimed at improving the efficiency of AI agents during search tasks. This method leverages a strategy of 'dreaming' about past attempts…
Problem The paper addresses a gap in the capability of artificial agents to operate autonomously and adaptively by leveraging biological principles. It emphasizes the need for embodied agents that can…
A recent report from BankInfoSecurity discusses findings from the Korean AI Benchmark, which has uncovered notable deficiencies in multilingual safety within AI systems. This benchmark aims to evaluate the performance…
Problem This work addresses the gap in robotic manipulation tasks that require long-term memory capabilities without introducing spurious correlations. The authors highlight the need for a memory system that can…
Problem This paper addresses the challenge of modeling articulated objects from sparse monocular views, a significant gap in the literature. The authors propose a solution to improve the understanding of…
Problem The paper addresses a significant gap in the capability of existing image generation and editing models, specifically the ability to specify an object's target color using any 24-bit hex…
Problem This paper addresses the gap in understanding the overclaiming propensity of frontier large language model (LLM) agents regarding task completion. The authors aim to quantify how often these agents…
Problem This work addresses a gap in understanding the effectiveness of individual components in coding harnesses for autonomous coding agents. The study is particularly relevant as it is presented as…
Problem The paper addresses a gap in the capability of self on-policy distillation (OPD) for multi-turn agents in reinforcement learning (RL). The authors highlight that existing methods are hindered by…
Problem This preprint addresses the inadequacy of safety evaluations for large language models, specifically focusing on the persistence and transformation of discriminatory content rather than its removal. The authors argue…
Problem The paper addresses a significant gap in the capability of Variable Length Action (VLA) policies, specifically the limitation imposed by a fixed action horizon in action chunking. This constraint…
Problem The paper addresses a gap in the capability of generative agents to produce verifiable and steerable video highlight outputs. This issue is particularly relevant in the context of sports…
Problem Disaggregated assessment of AI system performance across various domains is essential for understanding model efficacy. However, exhaustive testing on all possible scenarios is prohibitively expensive. This paper addresses the…
Problem This paper addresses the limitations of existing retrieval-augmented generation (RAG) systems, particularly their inability to effectively manage the multi-stage and stateful nature of support cases. The authors highlight that…
Problem The paper addresses a significant gap in the capability of falsifying specifications in cyber-physical systems (CPS) using traditional black-box search algorithms. Existing methods often struggle with the complexity and…
Problem The paper addresses a gap in the capability of large language model (LLM) driven retrieval-augmented generation (RAG) systems specifically for spreadsheets. The authors highlight the need for improved interpretability…
Problem This work addresses the challenge of identifying optimal steering parameters for modifying the behavior of large language models (LLMs) during inference. The authors propose a novel approach, Deep Noir,…
Problem Standard supervised fine-tuning (SFT) in reinforcement learning typically applies loss only to action tokens generated by the agent, neglecting the environment observations. This paper addresses this gap by proposing…
Problem Adapting large-scale vision-language-action models to specific deployment scenarios is challenging due to the limited coverage of out-of-distribution states and the ineffectiveness of imitation objectives. This paper addresses these issues…
Problem This preprint addresses a gap in the understanding of how AI impacts the sense of ownership in collaborative tasks. The authors highlight the need for insights into the dynamics…
Problem The paper addresses the inadequacy of Allen's interval algebra in modeling uncertain temporal information. Traditional interval algebra does not account for probabilistic relationships between time intervals, which is crucial…
