paper-with-me

홈 › Papers

TriAttention: Efficient Long Reasoning with Trigonometric KV Compression

2026-04-06 · Weian Mao, Xi Lin, Wei Huang, Yuxin Xie, Tianfu Fu, Bohan Zhuang, Song Han, Yukang Chen arxiv

Extended reasoning in large language models (LLMs) creates severe KV cache memory bottlenecks. Leading KV cache compression methods estimate KV importance using attention scores from recent post-RoPE queries. However, queries rotate with position during RoPE, making representative queries very few, leading to poor top-key selection and unstable reasoning. To avoid this issue, we turn to the pre-RoPE space, where we observe that Q and K vectors are highly concentrated around fixed non-zero centers and remain stable across positions -- Q/K concentration. We show that this concentration causes queries to preferentially attend to keys at specific distances (e.g., nearest keys), with the centers determining which distances are preferred via a trigonometric series. Based on this, we propose TriAttention to estimate key importance by leveraging these centers. Via the trigonometric series, we use the distance preference characterized by these centers to score keys according to their positions, and also leverage Q/K norms as an additional signal for importance estimation. On AIME25 with 32K-token generation, TriAttention matches Full Attention reasoning accuracy while achieving 2.5x higher throughput or 10.7x KV memory reduction, whereas leading baselines achieve only about half the accuracy at the same efficiency. TriAttention enables OpenClaw deployment on a single consumer GPU, where long context would otherwise cause out-of-memory with Full Attention.

📄 PDF Abstract BibTeX arXiv:2604.04921

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

CASK: Core-Aware Selective KV Compression for Reasoning Traces

2026-04-13 · Buseong Kim, Heejun Gwon arxiv

In large language models performing long-form reasoning, the KV cache grows rapidly with decode length, creating bottlenecks in memory and inference stability. Existing reasoning-oriented KV compression has mostly follow…

Minimal-Intervention KV Retention via Set-Conditioned Diversity

2026-05-14 · Libo Sun, Po-wei Harn, Peixiong He, Xiao Qin arxiv

KV-cache compression at small budgets is a crowded design space spanning cache representation, head-wise routing, compression cadence, decoding behavior, and within-budget scoring. We study seven mechanisms across these …

Mathematical Reasoning

Adaptive Mass-Segmented KV Compression for Long-Context Reasoning

2026-05-22 · Junzhe Yang, Xiaoyu Shen arxiv

The linear growth of the Key-Value (KV) cache is a critical bottleneck in long-form LLM inference. Existing KV compression methods mitigate this by evicting tokens based on importance scores. However, we show that their …

Mathematical ReasoningCode Completion

TRIGO: Benchmarking Formal Mathematical Proof Reduction for Generative Language Models

2023-10-16 · Jing Xiong, Jianhao Shen, Ye Yuan, Haiming Wang 외

Automated theorem proving (ATP) has become an appealing domain for exploring the reasoning ability of the recent successful generative language models. However, current ATP benchmarks mainly focus on symbolic inference, …

Automated Theorem ProvingBenchmarkingMathematical Reasoning

Think Less, Know More: State-Aware Reasoning Compression with Knowledge Guidance for Efficient Reasoning

2026-04-10 · Yi Sui, Chaozhuo Li, Dawei Song arxiv

Large Reasoning Models (LRMs) achieve strong performance on complex tasks by leveraging long Chain-of-Thought (CoT), but often suffer from overthinking, leading to excessive reasoning steps and high inference latency. Ex…

Mathematical Reasoning