All developments.

As the production of software passes into the hands of software itself, coding agents are increasingly commissioned to build agents fitted to their users' own workflows.

When solving complex spatial reasoning tasks, ordinary humans naturally revisit the scenes from task-relevant viewpoints and mentally reconstruct the spatial layouts from those viewpoints.

Computer-use agents can perform increasingly complex GUI tasks, yet unfamiliar software and workflows requiring human domain knowledge remain challenging.

Decoder hidden states from different layers of large language models (LLMs) often encode complementary information.

Continual post-training faces a dual challenge of acquisition and retention: models must learn new capabilities while preserving prior knowledge.

Continual learning requires models to accumulate past experience and transfer it to new tasks.

Coding agents are increasingly used to build programs from scratch as software engineers do rather than to resolve issues within a repository or to generate a repository from a given requirements document.

Refugee gig workers for tech companies don’t know how much they’ll be paid – if AI hasn’t already taken their jobs In a portable building in Kakuma, a refugee camp near Kenya ’s northern border with South Sudan , Grace u…

The story is a collaboration between MIT Technology Review and Aventine, a non-profit research foundation that creates and supports content about how technology and science are changing the way we live.

Joint-embedding predictive architectures (JEPAs) learn compact latent world models without pixel reconstruction.

Humans perceive far more in a scene than what is explicitly depicted: a single glance captures past causes and future trajectories; a quick peek determines if a vehicle can fit between two parked cars; a few seconds of v…

Continual learning (CL) seeks to adapt pretrained models while preserving prior capabilities.

Generating realistic human-object interactions (HOI) in complex 3D scenes requires two complementary capabilities: reasoning about interaction feasibility in the environment and synthesizing realistic human-object motion…

Knowledge distillation (KD) transfers knowledge from stronger Teacher models to weaker Student models, but most methods require training the Student parameters, thereby binding the distilled knowledge to a specific archi…

Data video is a media form that integrates data visualization with video narrative, widely adopted in news reporting and business analysis.

Adversarial games have driven advances from heuristic search to reinforcement learning, yet learning and adapting strategies from limited samples remain challenging.

Language model agents increasingly operate over long interaction streams, requiring online adaptation to distribution shifts through continual learning.

We propose BrickBench, a benchmark for agentic text-conditioned LEGO-set design.

When retrieved evidence contradicts an agent's prior beliefs, does it revise its answer, acknowledge uncertainty, or persist with an incorrect conclusion?

Egocentric world models predict first-person observations conditioned on an agent's actions, but most focus on a single agent.

Spatial reasoning remains a persistent weakness of vision-language models (VLMs), because RGB inputs do not directly provide geometric evidence.

In this paper, we argue that looped Transformers need their own residual connections to prevent performance degradation as the number of iterations grows.

In text-guided image editing, describing the desired change is often straightforward, but identifying the intended object or region can be cumbersome, especially when several objects look alike.

Text prompting is an indirect interface for poster generation, requiring users to encode inherently two-dimensional composition intent into a one-dimensional sequence of words.

The evaluation of coding agents generally strives to be as realistic as possible.

Skeleton-based reasoning prompting is a promising training-free approach for structuring LLM reasoning, but prior work largely assumes an English-centric setting.

Latent world models have shown a remarkable ability to predict future states and to plan in the real world.

A large language model can only use the text that fits in its context window, and it recomputes its internal key-value (KV) state for a prompt every time the prompt is sent.

Pixel-space diffusion models avoid the lossy VAE of latent models, which suggests an advantage on downstream tasks where fine-grained detail matters.

Methods that recover individual latent variables of the world, from nonlinear ICA to dictionary learning and causal representation learning, anchor the latents to observations through reconstruction, auxiliary supervisio…

Training and evaluating interactive language agents typically requires rich user interactions, yet collecting human feedback is expensive and difficult to scale.

Dual-process agents pair a fast policy with a slow deliberative model.

Large language model agents can improve across tasks by retaining reusable skills distilled from prior interactions.

Flow policies capture rich and diverse action distributions, and fine-tuning them with off-policy RL to improve beyond the demonstrations has drawn growing interest.

Multi-teacher on-policy distillation (MOPD) is used in two settings.

Test-time scaling (TTS) improves the reasoning capabilities of large language models by allocating additional inference computation.

Memory-augmented reinforcement learning strengthens LLM agents' ability to solve complex long-horizon tasks.

Much knowledge work produces new deliverables from files a workspace already holds, and LLM agents are beginning to take such work over.

Recent advances in multimodal foundation models have made them capable generalist physical agents for a range of manipulation tasks.

Dense visual text requires image generators to reproduce long strings across multiple regions with correct placement and legibility.

Real-time robot control demands enough visual history to infer motion and task progress, but processing that history can delay action.

Agentic systems use large language models (LLMs) to carry out concrete tasks.

Improving embodied agents often focuses on optimizing the underlying model through training, while the surrounding agent harness that controls planning, context, and tool use is typically engineered.

Multimodal large language models have become the dominant paradigm for visual understanding, but incur substantial costs by encoding inputs into dense, fixed-size patch tokens.

Automated research systems increasingly run LLM agents over long horizons, but more inference does not by itself produce more progress: agents replay growing histories, duplicate one another's work, or stop experimenting…

Typed decision models answer a declared question without generating text: a decision head returns a probability for each of the declared options in a single forward pass.

Reinforcement learning with verifiable rewards (RLVR) turns agent experience into learning signals primarily through scalar outcome rewards after interaction.

AI agents increasingly operate in environments where they can diagnose failures and improve through experience, yet existing evaluations largely measure what an agent can do at a fixed point in time rather than how effec…

A central challenge for scientific agents is to turn analytical experience into reusable expertise grounded in physical evidence.

Reconstructing an editable CAD model from a 3D shape remains a challenging engineering task.