Crossmodal matching in single-cell omics is essential for explaining biological regulatory mechanisms and enhancing downstream analyses. However, current single-cell crossmodal models often suffer from three limitations: sparse modality signals, underutilization of biological attributes, and insufficient modeling of regulatory interactions. These challenges hinder generalization in data-scarce settings and restrict the ability to uncover fine-grained biologically meaningful crossmodal relationships. Here, we present a novel framework which reformulates crossmodal matching as a graph classification task on Attributed Bipartite Graphs (ABGs). It models single-cell ATAC-RNA data as an ABG, where each expressed ATAC and RNA is treated as a distinct node with unique IDs and biological features. To model crossmodal interaction patterns on the constructed ABG, we propose Bi2Former, a biologically-driven bipartite graph transformer that learns interpretable attention over ATAC-RNA pairs. This design enables the model to effectively learn and explain biological regulatory relationships between ATAC and RNA modalities. Extensive experiments demonstrate that Bi2Former achieves state-of-the-art performance in crossmodal matching across diverse datasets, remains robust under sparse training data, generalizes to unseen cell types and datasets, and reveals biologically meaningful regulatory patterns.

With the development of large language models (LLMs), efficient inference through Key-Value (KV) cache compression has attracted considerable attention, especially for long-context generation. To compress the KV cache, recent methods identify critical KV tokens through static modeling of attention scores. However, these methods often struggle to accurately determine critical tokens as they neglect the temporal patterns in attention scores, resulting in a noticeable degradation in LLM performance. To address this challenge, we propose AttentionPredictor, which is the first learning-based method to directly predict attention patterns for KV cache compression and critical token identification. Specifically, AttentionPredictor learns a lightweight, unified convolution model to dynamically capture spatiotemporal patterns and predict the next-token attention scores. An appealing feature of AttentionPredictor is that it accurately predicts the attention score and shares the unified prediction model, which consumes negligible memory, among all transformer layers. Moreover, we propose a cross-token critical cache prefetching framework that hides the token estimation time overhead to accelerate the decoding stage. By retaining most of the attention information, AttentionPredictor achieves 13 KV cache compression and 5.6 speedup in a cache offloading scenario with comparable LLM performance, significantly outperforming the stateof-the-arts.

Long-form video understanding poses a significant challenge for video large language models (VideoLLMs) due to prohibitively high computational and memory demands. In this paper, We propose FlexSelect, a flexible and efficient token selection strategy for processing long videos. FlexSelect identifies and retains the most semantically relevant content by leveraging cross-modal attention patterns from a reference transformer layer. It comprises two key components: (1) a training-free token ranking pipeline that leverages faithful cross-modal attention weights to estimate each video token's importance, and (2) a rank-supervised lightweight selector that is trained to replicate these rankings and filter redundant tokens. This generic approach can be seamlessly integrated into various VideoLLM architectures, such as LLaVA-Video, InternVL and Qwen-VL, serving as a plug-and-play module to extend their temporal context length. Empirically, FlexSelect delivers strong gains across multiple long-video benchmarks - including VideoMME, MLVU, LongVB, and LVBench. Morever, it achieves significant speed-ups (e.g., up to 9 on a LLaVA-Video-7B model), highlighting FlexSelect's promise for efficient long-form video understanding.

Gating mechanisms have been widely utilized, from early models like LSTMs [1] and Highway Networks [2] to recent state space models [3], linear attention [4], and also softmax attention [5, 6]. Yet, existing literature rarely examines the specific effects of gating. In this work, we conduct comprehensive experiments to systematically investigate gating-augmented softmax attention variants. Specifically, we perform a comprehensive comparison over 30 variants of 15BMixture-of-Experts (MoE) models and 1.7B dense models trained on a 3.5 trillion token dataset. Our central finding is that a simple modification--applying an head-specific sigmoid gate after the Scaled Dot-Product Attention (SDPA)--consistently improves performance. This modification also enhances training stability, tolerates larger learning rates, and improves scaling properties. By comparing various gating positions and computational variants, we attribute this effectiveness to two key factors: (1) introducing non-linearity upon the low-rank mapping in the softmax attention, and (2) applying query-dependent sparse gating scores to modulate the SDPA output. Notably, we find this sparse gating mechanism mitigates'massive activation' [7], 'attention sink' [8], and enhances long-context extrapolation performance, and we also release related codes and models to facilitate future research. Furthermore, the most effective SDPA output gating is used in the Qwen3-Next models.

Large vision-language models (LVLMs) excel at multimodal understanding but suffer from high computational costs due to redundant vision tokens. Existing pruning methods typically rely on single-layer attention scores to rank and prune redundant visual tokens to solve this inefficiency. However, as the interaction between tokens and layers is complicated, this raises a basic question: Is such a simple single-layer criterion sufficient to identify redundancy? To answer this question, we rethink the emergence of redundant visual tokens from a fundamental perspective: information flow, which models the interaction between tokens and layers by capturing how information moves between tokens across layers. We find (1) the CLS token acts as an information relay, which can simplify the complicated flow analysis; (2) the redundancy emerges progressively and dynamically via layer-wise attention concentration; and (3) relying solely on attention scores from single layers can lead to contradictory redundancy identification. Based on this, we propose FlowCut, an information-flow-aware pruning framework, mitigating the insufficiency of the current criterion for identifying redundant tokens and better aligning with the model's inherent behaviors. Extensive experiments show that FlowCut achieves superior results, outperforming SoTA by 1.6% on LLaVA-1.57B with 88.9% token reduction, and by 4.3% on LLaVA-NeXT-7B with 94.4% reduction, delivering 3.2 speed-up in the prefilling stage.

Differential Transformer has recently gained significant attention for its impressive empirical performance, often attributed to its ability to perform noise canceled attention. However, precisely how differential attention achieves its empirical benefits remains poorly understood.

KV cache eviction has emerged as an effective solution to alleviate resource constraints faced by LLMs in long-context scenarios. However, existing token-level eviction methods often overlook two critical aspects: (1) their irreversible eviction strategy fails to adapt to dynamic attention patterns during decoding (the saliency shift problem), and (2) they treat both marginally important tokens and truly unimportant tokens equally, despite the collective significance of marginal tokens to model performance (the marginal information over-compression problem). To address these issues, we design two compensation mechanisms based on the high similarity of attention matrices between LLMs of different scales. We propose SmallKV, a small model assisted compensation method for KV cache compression. SmallKV can maintain attention matching between different-scale LLMs to: 1) assist the larger model in perceiving globally important information of attention; and 2) use the smaller model's attention scores to approximate those of marginal tokens in the larger model. Extensive experiments on benchmarks including GSM8K, BBH, MT-Bench, and LongBench demonstrate the effectiveness of SmallKV. Moreover, efficiency evaluations show that SmallKV achieves 1.75 2.56 times higher throughput than baseline methods, highlighting its potential for efficient and performant LLM inference in resource constrained environments.

As the length of input text increases, the key-value (KV) cache in LLMs imposes prohibitive GPU memory costs and limits long-context inference on resource constrained devices. Existing approaches, such as KV quantization and pruning, reduce memory usage but suffer from numerical precision loss or suboptimal retention of key-value pairs. In this work, Low Rank Query and Key attention (LRQK) is introduced, a two-stage framework that jointly decomposes full-precision query and key matrices into compact rank-r factors during the prefill stage, and then employs these low-dimensional projections to compute proxy attention scores in O(lr) time at each decode step. By selecting only the top-k tokens and a small fixed set of recent tokens, LRQK employs a mixed GPU-CPU cache with a hitand-miss mechanism where only missing full-precision KV pairs are transferred, thereby preserving exact attention outputs while reducing CPU-GPU data movement.

Rotary Positional Encodings (RoPE) have emerged as a highly effective technique for one-dimensional sequences in Natural Language Processing spurring recent progress towards generalizing RoPE to higher-dimensional data such as images and videos. The success of RoPE has been thought to be due to its positional equivariance, i.e. its status as a relative positional encoding. In this paper, we mathematically show RoPE to be one of the most general solutions for equivariant positional embedding in one-dimensional data. Moreover, we show Mixed RoPE to be the analogously general solution for M-dimensional data, if we require commutative generators - a property necessary for RoPE's equivariance. However, we question whether strict equivariance plays a large role in RoPE's performance. We propose Spherical RoPE, a method analogous to Mixed RoPE, but assumes non-commutative generators. Empirically, we find Spherical RoPE to have the equivalent or better learning behavior compared to its equivariant analogues. This suggests that relative positional embeddings are not as important as is commonly believed, at least within computer vision. We expect this discovery to facilitate future work in positional encodings for vision that can be faster and generalize better by removing the preconception that they must be relative.

We demonstrate that geometrically distinctive keys during LLM inference tend to have high attention scores. Based on the phenomenon we propose KEYDIFF, a training-free KV cache eviction method based solely on key similarity. Unlike other KV cache eviction methods, KEYDIFF can process arbitrarily long prompts within strict resource constraints and efficiently generate responses. We provide a theoretical basis for KEYDIFF by relating key diversity with attention scores. These results imply KEYDIFF can efficiently identify the most important tokens to retain. Notably KEYDIFF does not rely on attention scores, allowing the use of optimized attention mechanisms like FlashAttention. Under a strict memory allowance, we demonstrate the effectiveness of KEYDIFF for the Llama and Qwen model families by observing a performance gap of less than 0.04% with 8K cache budget ( 23% KV cache reduction) from the non-evicting baseline on LongBench for Llama 3.1-8B and Llama 3.2-3B. We also observe near baseline performance for Deepseek-R1-Distill-Llama-8B on the Math500 reasoning benchmark and decrease end-to-end inference latency by up to 30% compared to the other token-eviction methods.