3,698 items tracked: 33 reported by two or more outlets, the rest by a single source. Most-corroborated first.
Recent work on language-model adaptation has shown that single models can obtain informative training signals by evaluating their behavior in demonstrationor feedback-augmented contexts, with the help of a teacher networ…
Video interactive storytelling enables viewers to actively steer how a video unfolds.
When audio, video, and text disagree, accuracy alone does not show whether to acquire another source or abstain.
Vision-and-Language Navigation (VLN) has largely focused on a single agent following a single instruction, yet many real-world applications require teams of robots to tackle tasks beyond the capabilities of any individua…
As text-to-SQL systems move beyond demonstrations toward real-world deployment, ambiguity in user questions becomes a primary source of errors.
Recovering faithful 3D hand motion from video remains challenging due to frequent occlusions and incomplete visual observations, which make frame-wise pose estimates unreliable and temporally inconsistent.
Video-language models commonly assume that more temporal observations lead to more reliable reasoning.
Large-scale egocentric human interaction data is becoming an important source of physical supervision for embodied learning, yet video alone leaves the contact and pressure that characterize physical interaction unrecord…
On-policy distillation (OPD) strengthens language-model reasoning, yet whether students acquire new factual knowledge or compositional skill for multi-step reasoning remains unknown.
Conditional memory architectures such as DeepSeek Engram use input n-grams to look up learned embeddings, expanding the capacity of large language models (LLMs) with limited additional computation.
We introduce QuadTok, a novel framework for visual tokenization and autoregressive image generation.
We propose Tetris3D, a generative framework for single-image 3D scene reconstruction that recovers objects which are physically and geometrically coherent as a scene.
We introduce SWE-Game, a benchmark of 247 tasks grounded in 41 executable reference Godot games spanning 13 gameplay categories in 2D and 3D.
General-purpose agents increasingly write code, use tools, and complete complex digital tasks, raising the question of how far these capabilities carry into the physical world.
Highly compressed video autoencoders offer an effective way to accelerate video diffusion models, as the Diffusion Transformer (DiT) operates on far fewer tokens.
Prevailing multi-vector visual document retrievers store each page as about a thousand patch vectors, often in vector databases run by a third party.
Autoregressive video generation requires denoising the current frames while writing their key-value representations as context for future predictions.
Product-centric advertisement video generation aims to create promotional videos that preserve fine-grained product identity while presenting selling points through coherent multi-shot narratives.
We present WebFovea, a vision-based web agent that placed 2nd in the WebRetriever Challenge 2026 with a final score of 57.0 out of 100.
The architectural design of Large Language Models (LLMs) is shifting from traditional full-attention-only models to hybrid models, which combine different attention modules to improve long-context efficiency and performa…
Recently, approaches that leverage human video datasets for robot policy training have become increasingly prevalent.
Jennifer Neville did not want to go into computer science—but that’s exactly where she landed.
The rapid growth of 3D Gaussian Splatting (3DGS) research demands significant effort to reimplement papers before building on them.
A declared task can sometimes certify a reduced source before a downstream encoder, codebook, rate, distortion target, or optimizer is chosen.
Vision-language-action (VLA) policies trained on large datasets are capable within their training domains, yet they still fail to generalize to the variety of situations a robot meets in real-world deployment.
Speculative decoding accelerates autoregressive generation in large language models.
Proactive AI assistants continuously observe a user's activity and decide whether to provide new guidance or remain silent.
On-policy distillation (OPD) has been widely studied as a post-training method in which a student model obtains token-level supervision from a stronger teacher on its own rollouts.
As large language model (LLM) agents become increasingly autonomous, they may pursue task performance through deception, raising concerns about their reliable deployment.
A safe action is not necessarily a viable one.
Long-context inference and Retrieval-Augmented Generation (RAG) handle evidence selection at vastly different scales, from a single long prompt to an entire corpus.
Interactive video world models need to generate each video chunk efficiently while responding faithfully to user controls.
Site-specific fertilizer recommendation systems adapt nutrient advice to location, soil properties, crop type, and production targets, but scientific reuse is constrained when recommendation functions remain accessible m…
Vision-language models (VLMs) are increasingly trained to generate structured outputs like points and bounding boxes that downstream interfaces, agents, and robots can act on, yet safety alignment of this output channel…
Massive activations, residual-stream coordinates with magnitudes far larger than typical activations, are associated with attention sinks in transformers, but how their scale is regulated during training remains incomple…
Diffusion large language models dLLMs) have emerged as a promising alternative to autoregressive language models through bidirectional diffusion-based token generation.
Sparse attention is a primary approach to reducing the latency of diffusion transformers in long-sequence generation tasks, such as video and high-resolution 3D asset generation.
Realistic environment replicas are increasingly valuable for training and evaluating LLM agents, yet the original systems may be inaccessible or impractical to reproduce.
As agents take on long-horizon tasks, users shift from making individual decisions to overseeing autonomous execution.
Selective state space models (SSMs), such as Mamba, achieve strong per-token expressivity by making the time discretization step TildeΔ a learned function of the input.
Multimodal Large Language Models (MLLMs) have demonstrated remarkable potential in video understanding, yet their reliance on retrospective summarization and text-centric priors often limits their ability to bridge unobs…
Autoregressive video diffusion supports streaming generation and interactive control, but its KV cache grows continuously with the generated history.
Continuous asynchronous replanning is essential for real-time generative robot policies, but independent stochastic initialization can cause mode switching and inconsistent continuation across action chunks.
Multimodal imitation learning requires diverse executable futures under the same observation and consistent behavior across replanning cycles.
Bidirectional video diffusion models denoise entire videos in parallel, yet when trained on effectively unlimited in-distribution data from procedural generators, continue to violate physical laws and simple symbolic rul…
Long-horizon LLM agents are frequently implemented using sparse mixture-of-experts (MoE) models, yet the co-design of agentic behavior and MoE structures remains underexplored.
Multimodal Diffusion Transformers (MM-DiTs) jointly process visual and textual representations throughout generation.
Open-ended discovery requires learning from repeated attempts while continuing to explore directions whose value is not yet apparent.
Recent GPU kernel optimization agents retain what they learn in knowledge bases or as distilled skills.
Large language models (LLMs) are increasingly used as interactive agents and simulators, yet it remains unclear whether they can recover latent sequential structure beyond surface action frequencies.