The paper introduces Flash-dLLM, a framework for accelerating diffusion large language models (dLLMs) by addressing I/O bottlenecks in key-value (KV) caching and enabling efficient parallel decoding. It proposes an I/O-aware fused KV-cache kernel and a draft-and-verify decoding strategy that uses the dLLM itself for both tasks, improving speed and memory efficiency.
The chart will appear after repeat observations. The current metric comes from the source.
3 source votesReal observations only. History before source connection is not reconstructed.
Flash-dLLM improves inference speed and memory efficiency for diffusion LLMs by optimizing I/O in KV caching and enabling parallel decoding without additional models.
The paper introduces Document Retrieval-Aware Chunking (D-RAC), a method for efficiently processing enterprise documents by normalizing them into PDF and converting them into retrieval-optimized Markdown using a multimodal LLM. This approach preserves document structure, reduces token costs, and improves chunking efficiency compared to traditional methods.
2 days agoDaily PapersIntBMoE introduces a block-conditioned Mixture-of-Experts (MoE) framework that decouples participation, execution, and materialization by combining dense expert composition with sparse block execution. It uses a learned codebook for blocks and a hypernetwork to merge experts, achieving full participation while keeping execution and memory costs low. Experiments show improvements in image classification, language modeling, and sequential recommendation, with real-world deployment in AMap's recommendation system.
5 days agoDaily PapersThe paper introduces SELF-INDEX, a framework that enables an index to self-evolve without human intervention by autonomously diagnosing retrieval issues, revising index keys, and validating changes. It also uses a Query Simulator to proactively explore new retrieval demands, improving performance across various corpora and downstream applications.
6 days agoDaily PapersCodeMidas is an agentic pipeline that converts implemented functionality in open-source codebases into executable reinforcement learning (RL) environments using only source code as input. It generates 5,545 training tasks from 3,185 codebases across 23 languages and 15 domains, improving performance on multiple coding benchmarks when training MiMo-V2.5 with GRPO.
5 days agoDaily PapersThis paper introduces FAMOS, a feed-forward model that predicts movable-part segmentation and joint parameters from sparse, unordered partial point clouds. The model uses a Multi-state Articulation Transformer to aggregate articulation cues across multiple observations and an observed articulation span objective to supervise motion ranges of parts.
6 days agoDaily PapersThe paper introduces OmniVChat, a task involving native audio-visual dialogue where models process audio and video inputs directly without separate text. It presents OmniVChat-Studio for generating dialogues, OmniVChat-Bench for evaluation, and OmniVChat-RL for reinforcement learning, improving performance on both synthetic and real-world data.
5 days ago