REVIEW 5 cited by
NACL: A General and Effective KV Cache Eviction Framework for LLMs at Inference Time
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Large Language Models (LLMs) have ignited an innovative surge of AI applications, marking a new era of exciting possibilities equipped with extended context windows. However, hosting these models is cost-prohibitive mainly due to the extensive memory consumption of KV Cache involving long-context modeling. Despite several works proposing to evict unnecessary tokens from the KV Cache, most of them rely on the biased local statistics of accumulated attention scores and report performance using unconvincing metric like perplexity on inadequate short-text evaluation. In this paper, we propose NACL, a general framework for long-context KV cache eviction that achieves more optimal and efficient eviction in a single operation during the encoding phase. Due to NACL's efficiency, we combine more accurate attention score statistics in PROXY TOKENS EVICTION with the diversified random eviction strategy of RANDOM EVICTION, aiming to alleviate the issue of attention bias and enhance the robustness in maintaining pivotal tokens for long-context modeling tasks. Notably, our method significantly improves the performance on short- and long-text tasks by 80% and 76% respectively, reducing KV Cache by up to 50% with over 95% performance maintenance. The code is available at https://github.com/PaddlePaddle/Research/tree/master/NLP/ACL2024-NACL.
Forward citations
Cited by 5 Pith papers
-
PhyCheck: Fine-Grained Evidence-Grounded Dataset for Physical Law Understanding in Video-LLMs
PhyCheck is a 69,825-pair video QA benchmark that tests and improves Video-LLMs' ability to judge whether events obey physical laws, with fine-grained evidence questions and a context-sensitivity pilot.
-
MadaKV: Adaptive Modality-Perception KV Cache Eviction for Efficient Multimodal Long-Context Inference
MadaKV adaptively splits the KV cache budget by attention-head modality preference and compensates across layers, cutting cache memory by 80-95% and speeding decoding by 1.3-1.5x with small accuracy loss.
-
Structured Thoughts For Improved Reasoning And Context Pruning
Structured try/outcome SFT improves math reasoning by up to 8% over standard SFT and enables pruning ~85% of context with ~9% accuracy drop.
-
AQUA: Attention via QUery mAgnitudes for Memory and Compute Efficient Inference in LLMs
A training-free method that prunes low-magnitude dimensions of projected query/key vectors in attention, cutting dot-product cost by 25% with small benchmark degradation.
-
DAM: Dynamic Attention Mask for Long-Context Large Language Model Inference Acceleration
DAM derives per-layer and per-head attention masks from a calibration dataset and extrapolates them to long inputs, matching full-attention retrieval accuracy while reducing memory and compute.
Discussion (0). Sign in to comment.