daily curated papers
Learning representations of CAD models is a largely open problem. While 3D representation learning has flourished around point clouds and meshes, the native format of CAD - boundary representations BReps, which encodes exact parametric surfaces …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
In this paper, we study regret minimization in repeated games with adaptive opponents who can respond based on histories of play. The standard metric of external regret in online learning is known to fail to …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Existing autonomous driving datasets have enabled major progress, but fall short in sensor fidelity, map completeness, or geographic diversity. We present KITScenes Multimodal, a European dataset built around high-fidelity sensors and maps. Our fully synchronized …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Selection is a core operation in interactive image editing. To be practical, a user should be able to specify and disambiguate the desired selection region through either text or click-based interactions, and the system should …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Off-policy reinforcement learning of pretrained flow policies remains challenging due to the instability of optimization arising from the multi-step sampling process. Recently, Q-learning with Adjoint Matching (QAM) addressed this issue by reformulating into a memoryless …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Agentic LLMs with web search change the threat model for text anonymization: weak contextual cues can become cross-referenceable evidence for re-identification, yet those same details also carry downstream analytic value of the text. Existing defenses …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Large language models are increasingly deployed as coding agents, shifting safety from individual responses to action sequences. Existing benchmarks, however, primarily assess whether models refuse unsafe prompts, leaving impacts on stateful workspaces largely unexamined. We …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Multimodal Large Language Models (MLLMs) excel at 2D semantic understanding but lack intrinsic 3D awareness, resulting in representations that fail to maintain geometric and spatial consistency across video frames. Given the scarcity of large-scale 3D …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Vision-Language-Action (VLA) models leverage the rich world knowledge of pretrained vision-language models (VLMs) to enable instruction-following robotic manipulation. However, the structural mismatch between VLM semantic spaces and embodied control policies often hinders the learning of …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Code language models need repository-level context to resolve imports, APIs, and project conventions. Existing methods inject this knowledge as long inputs (retrieved through RAG or dependency analysis) or through per-repository fine-tuning and LoRA -- costly …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
A situated query like "where is Lin Wei?" often encodes more than its literal content: the user may also want to know whether Lin Wei is free, in a good mood, or worth interrupting now …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Large Language Models exhibit paradoxical fragility in fundamental arithmetic, implying a disconnect between internal computation and discrete output. By analyzing the residual stream geometry during multi-operand addition, we identify the Iso-Raw-Sum Trajectory (IRST), a geometric …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Benchmarks are fundamental for evaluating and advancing LLMs and MLLMs by providing standardized and explicit measures of performance. However, their construction is labor-intensive and hard to reuse, raising concerns about sustainability and scalability. Moreover, existing …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Large language models are increasingly used to simulate social media users and infer how individuals may respond to online discussions. However, it remains unclear whether these simulations reflect precise user-specific beliefs or whether they are …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
AI research often requires decisions before future evidence exists: which bottleneck to attack, which direction to pursue, or where a project should be positioned. We introduce ForeSci, a temporally controlled benchmark for evaluating whether LLM …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Financial AI agents often fail for a simple reason: they make users carry the complexity. A user must repeatedly restate goals, risk preferences, portfolio context, past judgments, and shifting market assumptions, while the agent answers …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Video generation models have made impressive strides in synthesizing visually compelling content, yet their outputs remain confined to the virtual domain. A natural question follows: how well do these models reflect the physical world when …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
In robotics systems, vast amounts of visual data are easily captured at high resolution using low-cost, low-power hardware. Yet, limited bandwidth and on-device compute resources prevent full utilization when transmitted via conventional codecs like JPEG/MPEG …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Temporal Grounding (TG) aims to localize video segments corresponding to a textual query. Prior research predominantly focuses on single-segment retrieval. Real-world scenarios, however, often require localizing multiple disjoint segments for a single query -- a …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Multimodal Large Language Models (MLLMs) have demonstrated significant achievements in general visual question answering (VQA) tasks. However, they remain brittle on mechanical engineering drawings, where high annotation density and weak domain knowledge, compounded by unreliable …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Large language models can reproduce training data, but existing memorization evaluations mostly measure whether models can be forced to do so, rather than whether they do so under ordinary use. We introduce PropMe, a propensity-aware …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Music recommendation systems typically treat songs as opaque tokens, relying on collaborative interaction histories which overlooks semantic or acoustic content. Prior work has explored LLM-augmented, multimodal, and text-enhanced approaches to sequential recommendation, and while some …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Video is temporally redundant: adjacent frames usually share most objects, background, and layout. Yet existing video multimodal large language models (video MLLMs) usually encode each sampled frame as an independent RGB image, causing visual tokens …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Agents are widely deployed as assistants over documents, tools, and code. However, they typically act only on explicit user requests, which surface only the problems the user has noticed, while many other important problems coexist …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Planning for real-world problems by language models often involves both world and user constraints, which may not be fully specified upfront and are progressively disclosed through interaction. However, existing benchmarks still underexplore adaptive planning under …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Automatic Speech Recognition (ASR) has become a key technology for human--AI interaction. However, code-switching ASR (CS-ASR) remains particularly challenging due to the severe scarcity of multilingual CS speech resources across diverse language pairs. Existing approaches …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Standard continuous-time generative models rely on monolithic architectures that must navigate vastly different signal regimes, from isotropic noise to intricate data distributions. While scaling model capacity improves performance, deploying a massive network uniformly across the …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
While household robots are often evaluated based on task completion, everyday domestic environments involve value-conflicting situations in which robots are expected to choose actions that prioritize other values than task success, such as human autonomy …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Training accurate medical image segmentation models requires large amounts of densely annotated data, which is costly and time-consuming to obtain. Semi-supervised learning (SSL) alleviates this by learning from both abundant unlabeled data and limited labeled …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Role-playing language agents (RPLAs) should play characters whose values and behavior evolve as the story progresses, not maintain a fixed persona. Existing benchmarks measure factual recall at a given chapter, not whether responses align with …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Prior work has shown that large language models (LLMs) can translate unseen or low-resource languages by undergoing continued training or even by encoding a grammar book in their context. However, both methods typically overfit specific …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Recent progress in Large Language Model (LLM) agents has enabled promising advances in automated data science. However, existing approaches remain fundamentally limited by their static action sets and lack of principled long-horizon context management, hindering …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Diffusion-based image editing has achieved strong visual fidelity under natural language instructions, yet most existing systems still operate at the level of surface instruction following, without reasoning about the implicit contextual constraints embedded in real …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Memory-augmented LLM agents tackle complex long-horizon tasks by recursively summarizing interaction trajectories into compact memory. However, existing approaches typically train these memory policies using outcome-based reinforcement learning, failing to localize where intermediate memory quality degrades …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Reinforcement Learning with Verifiable Rewards (RLVR) has recently emerged as the cornerstone for shaping the remarkable coding abilities of Large Language Models (LLMs). However, the scalability of RLVR is severely constrained by the scarcity of …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
System prompt optimization improves agent behavior without modifying the underlying model, yielding human-readable, model-agnostic instructions. Existing methods build a prompt agent that refines task agents' system prompts, yet leave the prompt agent's own system prompt …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Experience internalization converts contextual experience from past interactions into reusable parametric capability, offering a promising path toward continual learning in large language models (LLMs). While prior work has predominantly focused on single-iteration transfer, we discover …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Developing unified video generation and editing models capable of interpreting interleaved multimodal inputs is a promising yet challenging frontier field. Existing unified frameworks predominantly rely on massive models (typically 13B parameters or more) and incorporate …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Inference-time scaling has emerged as a critical avenue for enhancing Large Language Models' performance, yet real-world deployment is constrained by strict computational budgets. In this work, we formulate inference budget allocation as a global constrained …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
We propose world-language-action (WLA) models as a new class of embodied foundation models. WLA takes textual instructions, images, and robot states as inputs to jointly predict textual subtasks, subgoal images, and robot actions, conjoining the …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
World-action models (WAMs) jointly generate future video and robot actions through iterative diffusion, achieving strong performance on manipulation benchmarks but requiring tens of denoising steps, a cost that precludes real-time control. Step distillation has emerged …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Autonomous driving requires reasoning about how ego actions shape the evolution of the surrounding world. However, most end-to-end methods rely on direct state-to-action mappings, capturing correlations without explicitly modeling action-conditioned dynamics. Conversely, continuous-latent world models …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Large language models often improve reasoning by generating explicit chain-of-thought (CoT), demonstrating the importance of intermediate computation. However, textual CoT forces this computation through a discrete, serial, and communication-oriented token stream: each reasoning step must …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Video event prediction (VEP) requires models to infer unobserved future states from partial video evidence. Existing video MLLMs usually verbalize intermediate future reasoning in text space: once visual evidence is verbalized, fine-grained motion, geometry, and …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Processing video in vision-language models is expensive: each frame occupies hundreds of tokens, and inference cost scales with every frame and every repeated query. We introduce Video2LoRA, a method for parametric video internalization. A perceiver …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
We introduce VideoKR, the first large-scale training corpus specifically designed to strengthen knowledge- and reasoning-intensive video understanding. It comprises 315K video reasoning examples over 145K newly collected, CC-licensed, expert-domain videos. We develop a human-in-the-loop, skill-oriented …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Large language model (LLM) agents are increasingly applied to long-horizon tasks such as scientific discovery and machine learning engineering (MLE), where sustained self-evolution becomes a key capability. However, existing MLE agents suffer from inter-branch information …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
Inference-time skill augmentation provides a lightweight way to improve data-analytic agents by injecting reusable procedural knowledge without updating model parameters. However, discovering effective skills for data analysis remains challenging, as reliable supervision is expensive and …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
We study the personal camera roll visual question answering setting. In this setting, a conversational AI assistant can access a user's personal camera roll and retrieve relevant photos to answer queries, ranging from simple factual …
HUGGINGFACE_PAPERS daily curated papers
daily curated papers
On-policy distillation (OPD) supervises the student only in output space by matching next-token probabilities. This output-only paradigm has two limits: (1) sampling variance from Monte Carlo KL estimates over large vocabularies (e.g., Qwen's ~150k tokens) …
HUGGINGFACE_PAPERS daily curated papers