8 papers
DiffResearch: Accelerating Deep Research with Diffusion Language Models
Alan-Barsag Gazzaev, Alexey Gavrilov, Mikhail Mozikov +1
Deep research systems iteratively plan, gather evidence, synthesize, and revise, making end-to-end latency a key bottleneck for interactive use. Almost all production systems are b…
Hidden Messages Between Diffusion Language Model Agents: Causal Evidence for Latent Communication
Alan-Barsag Gazzaev, Alexey Gavrilov, Sergey Muravyov +1
Diffusion language models form answers through a sequence of revisable continuous states, but multi-agent systems usually discard those states and make agents communicate only afte…
AquaJEPA: An Action-Conditioned Multimodal JEPA Family for Underwater Robot Dynamics
Alan-Barsag Gazzaev, Alexey Gavrilov, Sergey Muravyov
Underwater robots rely on complementary sensors whose reliability changes abruptly with water visibility and vehicle motion. We introduce AquaJEPA, a sensor-configurable family of…
One Future, Every Robot: Label-Efficient Collective-State Prediction with Decentralized JEPA
Alan-Barsag Gazzaev, Alexey Gavrilov, Sergey Muravyov
Decentralized robots often need a common view of what their team is becoming, even though each robot sees different evidence and cannot rely on a central estimate or output-level c…
Where Quality Breaks in Compressed Short-Text Generation: Staged Bottleneck Localization
Alexey Gavrilov, Alan-Barsag Gazzaev, Sergey Muravyov
Compressed short-text generators can fail in two different places: the codec may discard information before generation starts, or the latent generator may produce weak codes. Witho…
PA-CoT: Profile-Adaptive Chain-of-Thought for Personalized Nutritional Consulting
Evgenii Garmashov, Nikita Kulin, Artur Khairullin +5
In health and nutrition consulting, widely used prompting methods pass the user profile as an unstructured block without a dedicated analysis step, leaving personalization as a cri…
From Monolithic to Modular: Segment-level Automatic Prompt Optimization
Nikita Kulin, Viktor Zhuravlev, Artur Khairullin +4
Automatic Prompt Optimization (APO) often rewrites prompts monolithically, which can improve one behavior while degrading others. We present SAPO, a segment-level APO method that d…
Mitigating Collaborative Semantic ID Staleness in Generative Retrieval
Vladimir Baikalov, Iskander Bagautdinov, Sergey Muravyov
Generative retrieval with Semantic IDs (SIDs) assigns each item a discrete identifier and treats retrieval as a sequence generation problem rather than a nearest-neighbor search. W…