Category
7,776 papers
HERO-MoE, Historical Expert ROuting with Scale-Preserving Fusion, a routing framework that injects historical routing priors into MoE routers by reusing detached, dense routing distributions collected from preceding MoE layers by reusing detached, dense routing distributions collected from preceding MoE layers.
Reinforcement learning (RL) for code agents often uses executable tests to provide binary rewards. With these rewards, Group Relative Policy Optimization (GRPO) assigns identical advantages to test-passing trajectories within each rollout group, overlooking differences in implementation quality and adherence to task re...
A new concept for designing corporate knowledge bases is proposed that integrates the probabilistic vector space of a corporate vocabulary, reflect-ing industry specifics, subject focus, terminology, and culture, with traditional ontological modeling, and enables the efficient extraction of knowledge from accumulated c...
Results show why reflection strategies should be assessed separately on final performance, parent improvement and calibration-to-test transfer.
Reach audiences
Advertise in front of researchers, engineers, and readers.
Direct Preference Optimization (DPO) aligns language models by optimizing over sequence-level sums of token-wise implicit reward differences. However, we identify a pervasive pathology in this formulation: a disproportionately small subset of high-frequency token types dominates cumulative sequence scores while appeari...
Overall, the results show that targeted feature quality depends jointly on the model signal and estimator, with detection and intervention capturing complementary properties.
LT-OPD, a training framework for extreme visual-token reduction, is proposed and it is shown that on-policy learning can substantially recover capabilities lost to extreme visual-token reduction.
Agents that answer questions from compressed or retrieved memory must recognize when the evidence a query needs is no longer in memory. Benchmarks for this task usually create insufficient-evidence examples by deleting supporting passages. We show that this construction leaks the label through memory size: on MuSiQue,...
Language models are now routinely graded by other language models. If post-training makes a model's own writing more predictable, it may also teach the same model, acting as a judge, to reward predictable writing, so that progress on creativity would be invisible to automated evaluation. We follow two open model famili...
Gate thresholds are implemented as couplings to a constant the network builds for the distribution it is reading, with the bias parameters contributing little; how much of that constant is carried in the stream and how much in parameters depends on the architecture.
Only on MathNet-Retrieve could an inversion, benchmark score up and real retention down, to one edit of a training file be pined, and three trained models are released.
Retrieval-Augmented Generation (RAG) introduces a specific failure mode in discrete diffusion language models: when retrieved context contradicts parametric knowledge, the iterative denoising process becomes a visible battleground between competing knowledge sources. We identify temporal semantic divergence as an obser...
A new computational approach reveals how cells communicate to assemble into tissues. This could accelerate stem cell engineering, improve disease models, and inform therapies.
Zero-knowledge proofs can verify computing results without rerunning workloads or exposing sensitive data. Magnetar helps bring them into the mainstream with support for standard programming languages, major speed gains, and quantum-resistant security. The post Proof-based general-purpose computing with Magnetar appeared first on Microsoft Research.
A new language-processing tool could help identify the highest-risk individuals from natural language, enabling swifter interventions.
A Blog post by Liquid AI on Hugging Face
Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking are our most advanced live dialogue models yet, built for natural conversation.
A new method, called CW-Net, translates the reasoning process of an autonomous vehicle’s AI system into understandable concepts that explain its behavior.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
