AI research papers.

Reinforcement learning (RL) for dexterous manipulation must discover finger-object contacts and then control the object precisely; the action noise that serves the first goal can interfere with the second.

Repeated training on model-generated data can degrade later models.

LLM agents generate intermediate reasoning and actions token by token, making extended interactions slow and computationally expensive.

Representation alignment (REPA) accelerates diffusion transformer training, but its alignment targets are almost exclusively semantic encoders such as DINOv2 and CLIP.

AI-generated content, often called AI slop, is increasingly common everywhere, particularly in academia.

Streaming world models should render a place consistently across repeated visits.

Vision-Language Models (VLMs) have shown strong multimodal reasoning capabilities, yet whether they truly capture the physical consistency underlying real-world dynamics remains unclear.

Multi-reward reinforcement learning trains large language models to satisfy multiple behavioral objectives simultaneously.

Recent video-based world models pair the scalability of autoregressive (AR) prediction with the visual quality of diffusion models.

Large Language Models (LLMs) have been used for novel discovery of algorithms, theorems, drugs, and other tasks through the use of harnesses that prompt an LLM to iteratively optimize an objective.

World models enable agents to plan by predicting future states of the environment, but their predictions can become unreliable when test-time dynamics differ from those seen during training.

Personalized image generation has remained narrowly focused on conditional synthesis from curated visual exemplars, rather than capturing who a user is.

Video generation has rapidly progressed from short, low-quality clips to high-resolution, long-duration sequences with complex spatiotemporal dynamics.

Dubbing quality control requires a reference-free judge that can determine whether a candidate text line matches a speaker's visible articulation in both content and timing, using only silent video and text because dubbe…

Many Indian banks serve customers through FAQ pages and scripted chatbots.

Biology doesn't operate in silos, and neither should the AI representation of it.

Coding agents can now write, run, and debug programs with little human help.

arXiv:2605.01352v2 Announce Type: replace-cross Abstract: Graphics and compute increasingly share a single GPU in embodied AI simulators, AI-enabled games, and VR systems.

arXiv:2609.07008v2 Announce Type: replace Abstract: Multi-head latent attention (MLA) exposes many logical query heads through one packed latent KV stream.

arXiv:2609.31652v1 Announce Type: cross Abstract: We present Open-Qwen-Music, an open reconstruction of Qwen-Music and a fully specified research system for text-to-music generation that couples LLM-based semantic compos…

arXiv:2609.31770v1 Announce Type: cross Abstract: General-purpose multimodal agents can write robot-control programs, but repeated exploration and model-mediated action selection can make execution slow.

arXiv:2609.32390v1 Announce Type: new Abstract: The July 2026 intrusion into Hugging Face production infrastructure showed how reward hacking can become an external cybersecurity incident when a capable agent encounters…

arXiv:2609.34180v1 Announce Type: new Abstract: How much does the decision readout matter when video-derived textual evidence is held fixed?

Latent reasoning has advanced multimodal reasoning through a two-stage training paradigm: (1) a helper image is encoded into latent tokens to teach visual chain-of-thought during a supervised fine-tuning (SFT) stage, and…

Recent multi-agent LLM systems increasingly combine heterogeneous models for specialized agent roles.

Coding agents are beginning to move beyond purely digital tasks to tackle physical-world challenges, particularly in robotics.

Existing online video segmentation methods struggle to track objects in long, complex videos with long-term occlusions.

Estimating mutual information (MI) from samples is a central objective in a variety of scientific fields.

An assistant that serves the same user over a long horizon has to answer from what that user has revealed: which preferences still hold, which were revised, and which constraints apply now.

On-policy distillation (OPD) trains a student to match the teacher's next-token distributions on the student's own trajectories and has yielded substantial empirical gains.

Recurrent neural networks (RNNs) compress the historical context into a memory state of fixed size, thus allowing for constant-time inference.

Pretrained transformers use little of their depth to follow references in context.

Transferring human motion to humanoid robots requires adapting the demonstrated motion to the robot morphology while preserving interactions with the environment.

Historical street-view imagery records urban evolution, but uneven coverage leaves substantial gaps in historical records.

Looped language models (LoopLMs) increase computational depth through parameter sharing, offering a path to scale inference computation without adding parameters.

Process reward models (PRMs) have become a key component for LLMs, as their step-level feedback supports both post-training and test-time reasoning.

Language-model agents are usually trained by reinforcement learning from one reward per episode, and privileged self-distillation enriches it by letting the same policy, given a skill, teach its skill-free self through t…

Long-context reasoning in large language models incurs substantial computation and memory costs.

Recurrent reasoning models have attracted growing attention for scaling test-time computation, typically by iteratively refining latent states with shared parameters.

The strong generalization performance of on-policy post-training paradigms has motivated studies of their parameter update behaviors.

Embodied agents now take on ever longer tasks.

Manipulation behaviors vary widely across objects and scenes, but they share a small set of reusable skills, and planning with these skills helps embodied agents generalize to new tasks.

Agents increasingly build on code written by other agents, and they reimplement rather than reuse, growing the codebases later agents must work in.

Group Relative Policy Optimization (GRPO) is widely used to train reasoning language models, where it computes advantages by centering and normalizing rewards across rollouts of the same prompt.

Recent efforts to scale tool-use post-training have largely centered on the synthesis of executable environments, which constitute only one component of a broader agentic interaction system comprising the environment, ta…

Reproducibility is essential for scientific research, yet prior work shows that LLM outputs vary with hardware and batching.

Manipulating next-token probabilities during generation can bypass the safety alignment of large language models.

System One models such as Jev offer an efficient alternative to generative language models for tasks that require decisions rather than open-ended responses.

Few-step streaming audio--video generation requires both causal modeling and step distillation, yet standard training recipes face two context-related challenges.