Research
New papers and findings, translated into plain language.
Research
Scaling up transformers has produced models with broad capabilities, deployed across heterogeneous hardware and fluctuating workloads. Yet, model families offer only a few fixed sizes (e.g., S, B, L), forcing users to...
arXiv · 17h ago
Research
We introduce Continuum Attention (CA), a causal Transformer attention operator that replaces discrete key and value lookup with integration over a continuous, piecewise linear memory curve. Each query induces an...
arXiv · 17h ago
Research
Long contexts enlarge key--value (KV) caches, increasing memory use and attention costs during generation. Compressing these caches must preserve information for unseen questions. We introduce **Latent Compaction**,...
arXiv · 17h ago
Research
We introduce a novel method for verifying Lean formalizations that goes beyond LLM-judged faithfulness. Unlike most verification methods, which rely on LLMs or trained models, our method combines LLM-based semantic...
arXiv · 17h ago
Research
We introduce an approach for self-supervised representation learning designed to be used as a tokenizer for generative modeling. Our objective supervises a representation loss by factorizing the pointwise mutual...
arXiv · 18h ago
Research
Recent advances in robot learning have driven rapid progress in humanoid control, enabling increasingly agile behaviors in the real world. However, humanoid control networks are typically obtained through conventional...
arXiv · 17h ago
Research
Mobile Manipulation (MoMa) is a cornerstone of autonomous robotic systems, yet real-world fine-tuning remains a formidable challenge because of high-dimensional action spaces and intricate coupling between heterogeneous...
arXiv · 17h ago
Research
Training language agents is bottlenecked by the high cost of long-horizon rollouts, yet standard agentic reinforcement learning (RL) learns only from a terminal scalar reward, leaving the supervision in intermediate...
arXiv · 17h ago
Research
An agentic market is a decentralized serving-time market where AI agents bid to claim each user prompt. An agent's bid is based on its expected value for the prompt, which is the expectation of the reward for its...
arXiv · 17h ago
Research
Current approaches for steady partial differential equations surrogate models on unstructured meshes apply a single mixing mechanism to the whole solution. This paper proposes an architecture that partitions the...
arXiv · 17h ago
Research
Computer-use agents have demonstrated strong capabilities in interpreting screens and executing keyboard and mouse actions. However, games, with their visually rich worlds and diverse mechanics, remain challenging and...
arXiv · 17h ago
Research
Post-training binds the competence of an LLM agent to the interface it was trained on. Semantics-preserving perturbations of that interface, such as renaming an action or changing the notation, leave the underlying task...
arXiv · 17h ago
Research
On-policy distillation (OPD) trains a student model on its own rollouts with token-level supervision from a teacher, assuming that the teacher can reach a correct answer from any prefix the student writes. We call this...
arXiv · 17h ago
Research
Brazilian Portuguese remains underserved by open-source language models, and the few that are safety-aligned rarely document what did not work or validate comparisons with statistical significance. We present...
arXiv · 17h ago
Research
Automated research with LLM agents has shown promising results in various domains. Through iterative refinement and scored verifiers, LLM agents are capable of discovering new ideas. There are different ways to scale...
arXiv · 17h ago
Research
3D object composition is pivotal to digital content creation and robot simulation, while also serving as a critical testbed for physical intelligence in foundation models. Previous methods either leverage large language...
arXiv · 17h ago
Research
Understanding the social intelligence of large language models (LLMs) is increasingly important for applications involving human-AI interaction and multi-agent environments. Social intelligence comprises two aspects:...
arXiv · 17h ago
Research
Potato is an important food-security crop in Ethiopia, but its production is affected by foliar diseases, particularly early blight and late blight. This study develops a lightweight convolutional neural network (CNN)...
arXiv · 17h ago
Research
Multimodal code intelligence connects the symbolic world of programs with the visual world their execution creates. Interest in this capability is burgeoning, as models are increasingly asked to produce charts,...
arXiv · 18h ago
Research
Relational deep learning has emerged as a powerful paradigm for learning end-to-end predictive models over relational databases by representing them as temporal heterogeneous graphs. Recent advances based on graph...
arXiv · 18h ago
Research
The full TechCrunch Disrupt Stage lineup revealed, featuring Max Hodak, Mark Wahlberg, Benchmark partners, and more. Register now to save up to $100 on your pass and get a second at 50% off.
TechCrunch · 2d ago