Long chain-of-thought (CoT) trajectories in large language model (LLM) reasoning cause severe inference bottlenecks due to rapid key-value (KV) cache growth. Current decoding-time compression methods mitigate this issue via token eviction, but typically assume a uniform budget distribution across all layers and heads.
The paper introduces Random Attention, a method that evicts KV cache entries uniformly at random within each attention head while preserving the prompt. Experiments on four models and six reasoning tasks show that this simple strategy matches the performance of the best existing eviction methods and achieves 32‑43% higher throughput in vLLM deployments. The authors explain that the prompt is the most fragile cache component and that reasoning traces are redundantly stored across text and attention heads, making a selection score unnecessary.
arXiv:2606. 11164v1 Announce Type: new Abstract: Long chain-of-thought (CoT) trajectories in large language model (LLM) reasoning cause severe inference bottlenecks due to rapid key-value (KV) cache growth.
arXiv:2608. 04771v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) excel on complex tasks through long chain-of-thought (CoT) reasoning, but their lengthy intermediate steps cause severe overthinking that inflates inference cost.
arXiv:2605. 06675v2 Announce Type: replace Abstract: Large language models cache all previously computed key-value (KV) pairs during generation, and this KV cache grows linearly with sequence length, making it a primary memory bottleneck for serving.
The paper introduces iS-KV, an online low‑rank KV‑cache compression technique that uses block‑incremental SVD to manage memory during long‑horizon autoregressive decoding. Unlike token‑eviction methods, iS‑KV retains all positions in a compact representation by keeping a recent window exact and incrementally folding older states into bounded‑rank bases, synchronizing coordinates as the basis evolves. Experiments on DeepSeek‑R1‑Distill‑Llama‑8B and Qwen3‑8B show that iS‑KV achieves high accuracy (82.6% and 89.2% respectively) while providing 4.06‑fold and 5.64‑fold compression, outperforming token‑eviction baselines under matched memory budgets.
The paper introduces ReadKV, a query‑adaptive quantization method for KV caches that stores keys and values in progressive codes, allowing each decoding query to allocate different precision prefixes for keys and values. By optimizing a calibrated distortion objective under fixed budgets, ReadKV achieves near‑optimal allocation and demonstrates that query‑dependent access can outperform any query‑independent reader at the same read budget. Experiments on six base models show that reading only four bits on average from an eight‑bit cache increases perplexity by at most 0.66% while reducing logical reads and retained capacity compared to a 16‑bit cache, and it achieves lower latency than existing codecs on an NVIDIA A10G.
AbridgeKV compresses long chain‑of‑thought outputs by encoding completed text blocks into the KV cache of specially trained pause tokens, freeing the original cache space. The method uses pause tokens positioned within the block, allocates cache budget per layer based on entry value, and compresses large blocks. Experiments on 446 competition problems show that with only a quarter of the original KV cache, AbridgeKV matches the accuracy and output length of a full‑KV model, while enabling faster inference and a 3.5× smaller cache per sequence.
The paper introduces Random Attention, a method that evicts KV cache entries uniformly at random within each attention head while preserving the prompt. It demonstrates that this simple strategy matches or surpasses more complex eviction schemes across four models and six reasoning tasks, achieving 32‑43% higher throughput in vLLM deployments. Experiments reveal that the prompt is the most fragile cache component and that redundancy in the reasoning trace across text and attention heads protects against random eviction, eliminating the need for a selection score.
Long-context large language model (LLM) inference is increasingly constrained by the memory footprint and decoding cost of key-value (KV) caches, limiting sustainable deployment on resource-constrained hardware. Existing KV cache eviction methods typically apply heuristic token scoring over all heads in GQA-based LLMs.
Large Reasoning Models (LRMs) excel on complex tasks through long chain-of-thought (CoT) reasoning, but their lengthy intermediate steps cause severe overthinking that inflates inference cost. KV-cache compression is a common solution, yet existing reasoning-oriented methods apply a uniform policy across the trajectory and judge compression only by what it removes from the cache.
The paper introduces EpiKV, an epiphany‑aware key–value cache eviction strategy that avoids using the attention matrix. It leverages hidden‑state shifts and recent query–key relevance to rank cached tokens, matching or surpassing the performance of existing attention‑based eviction methods while remaining compatible with fast inference kernels. Experiments on multiple benchmarks show that EpiKV improves inference throughput without sacrificing accuracy.