3,698 items tracked: 33 reported by two or more outlets, the rest by a single source. Most-corroborated first.
arXiv:2605.30638v2 Announce Type: replace-cross Abstract: We introduce Score Broadcast and Decorrelation (SBD), a principled framework for broadcast-based credit assignment for general families of differentiable losses.
arXiv:2606.06539v2 Announce Type: replace-cross Abstract: Forward-Forward (FF) learning [Hinton, 2022] replaces backpropagation with strictly layer-local goodness updates.
arXiv:2606.22272v2 Announce Type: replace-cross Abstract: Applying pre-trained language models to new domains through full fine-tuning is computationally expensive and prone to catastrophic forgetting.
arXiv:2606.22976v2 Announce Type: replace-cross Abstract: In this paper, we use random walks on graphs as a verifiable sandbox for studying parallel sampling strategies in masked diffusion models (MDMs).
arXiv:2606.24967v2 Announce Type: replace-cross Abstract: Hydraulic-conductivity inversion is ill posed: even complete head observations can leave structural ambiguity.
arXiv:2606.25879v2 Announce Type: replace-cross Abstract: Computational reproducibility remains difficult despite being central to scientific research.
arXiv:2607.17790v3 Announce Type: replace-cross Abstract: Egocentric devices, such as wearable front-facing cameras, provide a unique perspective for capturing the continuous interaction between a human viewer and the su…
arXiv:2607.19479v2 Announce Type: replace-cross Abstract: Existing teleoperation systems are often tailored to specific robot hardware and task domains, limiting their scalability and adaptability.
arXiv:2607.25970v2 Announce Type: replace-cross Abstract: RL for code correctness is now established: have the model generate a program, run it against hidden test cases, and reward solutions that pass.
arXiv:2607.28627v2 Announce Type: replace-cross Abstract: Long visual contexts challenge vision-language models: performance degrades as the number of distractors grows, and processing all tokens at once can exceed GPU m…
arXiv:2608.13368v3 Announce Type: replace-cross Abstract: This preliminary technical report presents a framework for sign language video synthesis using a loss-guided multi-expert Generative Adversarial Network (GAN) to…
arXiv:2608.23986v3 Announce Type: replace-cross Abstract: Large language model providers are compute constrained, and a common response to congestion is to degrade service.
arXiv:2608.30317v2 Announce Type: replace-cross Abstract: Dynamic origin-destination (OD) matrix estimation calibrates time-dependent input demand for simulations to reproduce observed link flows.
Multiplayer world models must generate independently controlled views with consistent representations of both players and their shared environment.
World action models (WAMs) aim to answer a coupled physical question: given a task instruction, what motion should the robot execute, and how will that motion change the surrounding world?
Multimodal large language models (MLLMs) struggle with spatial, embodied, physical, and temporal reasoning.
Large language model (LLM) assistants are being deployed to clinicians across health systems, and judgments about their readiness rest largely on benchmark scores, most of them derived from examination questions or curat…
Large Language Models (LLMs) increasingly ship with explicit "thinking modes", yet their counterpart, "no-thinking", has received far less attention.
Most image generation models rely on uniform tokenization, allocating the exact same computational budget to equally-sized image patches.
Large foundation models have accelerated progress toward general-purpose agents that interact with humans and other agents through language and multimodal signals.
Flow matching has emerged as a scalable paradigm for training high-quality generative models, but sampling from the learned probability flow requires many network evaluations.
A latent recurrent language model applies a shared block of layers repeatedly, so one checkpoint can trade inference compute for accuracy by changing the loop count.
Speculative decoding accelerates large language model inference by using a lightweight drafter to propose multiple tokens that are subsequently verified in parallel by the target model.
Computer-use agents (CUAs) are rapidly improving, yet synthesizing diverse and reliable software environments at scale remains challenging.
Humans effortlessly react to the world; they adeptly return a tennis shot, catch a falling cup, step aside from a passing cart, or pass a box tossed during truck unloading.
A 4D audio-visual scene comprises a video, the dynamic geometry it depicts, and the sound sources that populate it, each with its own position and trajectory.
Static coding benchmarks are fixed at release.
Online reinforcement learning (RL) fine-tunes diffusion models with black-box rewards, but reverse-process methods such as Flow-GRPO are limited in efficiency by poor credit assignment, while forward-process methods such…
Reward modeling often requires jointly representing and reasoning over multiple evaluation criteria, yet verbalizing this process token by token can incur substantial inference cost.
Most robot policies keep a model in the control loop: a VLA maps observations to actions, and an Agent Harness, such as Agent-as-Policy or Harness VLA queries a VLM for decision making at run time.
Multimodal large language models (MLLMs) are rapidly evolving toward continuous audio--visual reasoning, creating an urgent need for evaluations that expose their capability limits.
Single-image, multi-image, and video deep research require different visual operations but share a workflow of visual grounding, external retrieval, and fact composition.
Large language model (LLM) coding agents have advanced test generation across diverse programming tasks.
Existing spatial reasoning benchmarks mainly test spatial perception: reading off relations already visible in the input.
Skill-augmented agents improve sample efficiency by distilling successful trajectories into reusable strategies.
Spatial coding agents significantly improve spatial reasoning in Multimodal Large Language Models (MLLMs) by using external tools to generate verified execution traces.
Large language model (LLM) routing distributes inference work across different models, advancing the cost-quality Pareto frontier of LLM serving.
Generating an egocentric video from a single exocentric recording is a challenging case of novel view synthesis, as the two cameras share little overlap and much of the target view is unobserved.
We present DreamTrue, a multi-view, cross-embodiment robot world model for action-faithful and physically plausible video prediction.
General-purpose service robots need navigation systems that can handle diverse human requests in unfamiliar environments, combining task generality with scene generality.
Large Multi-modality Models (LMMs) have made significant progress in visual understanding and generation, but still face challenges in visual editing, particularly in following complex instructions, preserving appearance…
Geometrically faithful and functional articulated 3D assets are essential for real-to-sim robot manipulation, where policies trained in simulation must transfer to physical objects.
Dense retrievers have made significant progress on text and image corpora, but whether these capabilities extend reliably to mixed corpora containing text, image, and fused text-image documents remains unclear.
The memory-bound nature of the decoding stage of large language model (LLM) inference incurs significant latency.
Interactive virtual worlds allow agents to learn through exploration and interaction.
Epistemic status: a curiosity I chased for an afternoon.
OpenAI Group PBC has published 722 math papers that it generated using an unreleased artificial intelligence model.
The research found that important guardrails for teens using ChatGPT often failed.
Anthropic's new Claude Haiku 5.5 crushes its predecessor in benchmarks, jumping from 15.7 to 72.4 percent on the OSWorld computer use test.