This paper introduces Flex-Forcing, a framework for video generation that enables a model to operate under both bidirectional and autoregressive generation regimes, achieving better video quality and…

This study systematically evaluates a wide range of chunking methods for Retrieval-Augmented Generation (RAG) to assess their effectiveness and highlight the overlooked challenges associated with chun…

This paper evaluates the effectiveness of cluster-based semantic chunking compared to fixed-size and recursive chunking in Retrieval-Augmented Generation systems using the Retrieval Augmented Generati…

Soft-NBCE introduces soft entropy-weighted chunk fusion to overcome the semantic fragmentation caused by hard chunk selection in long-context LLMs, significantly improving performance on multi-hop ben…

The paper introduces an efficient, novel algorithm for incremental Byte Pair Encoding (BPE) tokenization that processes input text prefix by prefix, achieving significant speedups and enabling streami…

The paper proposes lightweight fine-tuning as an alternative to pooling-aware training for reducing vector count in late interaction models, achieving broad gains and high compression rates.

The paper introduces ChunkGroupSHAP, a listwise Shapley method that clusters semantantly related chunks into shared cross-document features for dense semantic ranking.

Murmur is an efficient inference system for long-form ASR that resolves the accuracy-latency trade-off by optimizing both inter-chunk processing and intra-chunk attention mechanisms.

AnyEdit++ introduces a structure-aware framework that uses Bayesian Surprise to adaptively segment long-form knowledge, significantly improving the coherence and accuracy of knowledge editing in LLMs.

The paper introduces CFGzip, an offline token space compression technique that significantly reduces the computational overhead of constrained decoding, making complex grammar enforcement feasible at…

The paper proposes SLAT, a segment-level adaptive trimming framework, which efficiently reduces redundant reasoning in large language model CoT outputs by selectively suppressing segments with low mar…

This paper introduces structured cut sets, a novel preprocessing technique for Maximum k-Cut, and extends existing techniques from Maximum Cut. The rules are optimality-preserving and yield significan…

The paper proposes SubFit, a novel compression technique that achieves superior LLM compression by replacing non-contiguous, submodule-level components (Attention and FeedForward) with lightweight res…

The paper introduces DocFormBench, a new benchmark for content-aware document formatting, and proposes DocFormFlow, a workflow that improves formatting accuracy and efficiency by decoupling target loc…

The paper introduces Self-Conditioned Positional HNSW (SCP-HNSW), a method that modifies chunk embeddings and retrieval process to mitigate redundant evidence retrieval from overlapping document chunk…

This paper proposes Head-Chunked Multi-Stream Pipeline (HCMS) to exploit the computational independence of multi-head attention and achieve fine-grained communication-computation overlap, resulting in…

DominoTree, a new best-first draft tree method for LLM inference, achieves up to 6.6x speedup over autoregressive decoding and the highest mean accept length, using a GPU-native CUDA-graph builder.

The paper proposes InSemRAG, an enhanced RAG framework that improves retrieval accuracy and knowledge integrity by incorporating intent-aware retrieval and semantics-preserving chunking, achieving sta…

The paper assembles a corpus of 102 papers on edge bundling and derives a task taxonomy for reasoning about bundled visualizations.