conftrace_

inference optimization

219 papers

Explore in graph

Co-occurring keywords

large language model (13587) model compression (3302) efficient computing (781) kv cache (74) speculative decoding (111) language model (4599) inference efficiency (248) inference acceleration (244) efficient inference (233) memory efficiency (261)

Papers

Fuzzy Speculative Decoding for a Tunable Accuracy-Runtime Tradeoff ACL 2025
Revisiting the Test-Time Scaling of o1-like Models: Do they Truly Possess Test-Time Scaling Capabilities? ACL 2025
HARP: Hesitation-Aware Reframing in Transformer Inference Pass NAACL 2025
Accurate KV Cache Quantization with Outlier Tokens Tracing ACL 2025
Accelerating Diffusion Transformer via Gradient-Optimized Cache ICCV 2025
DAST: Difficulty-Adaptive Slow-Thinking for Large Reasoning Models EMNLP 2025
Speculative Sampling via Exponential Races ACL 2025
TopV: Compatible Token Pruning with Inference Time Optimization for Fast and Low-Memory Multimodal Vision Language Model CVPR 2025
Inference Compute-Optimal Video Vision Language Models ACL 2025
MiniKV: Pushing the Limits of 2-Bit KV Cache via Compression and System Co-Design for Efficient Long Context Inference ACL 2025
ProCut: LLM Prompt Compression via Attribution Estimation EMNLP 2025
Data-Free Group-Wise Fully Quantized Winograd Convolution via Learnable Scales CVPR 2025
LAMB: A Training-Free Method to Enhance the Long-Context Understanding of SSMs via Attention-Guided Token Filtering ACL 2025
FBQuant: FeedBack Quantization for Large Language Models IJCAI 2025
BigMac: A Communication-Efficient Mixture-of-Experts Model Structure for Fast Training and Inference AAAI 2025
Treasures in Discarded Weights for LLM Quantization AAAI 2025
C3oT: Generating Shorter Chain-of-Thought Without Compromising Effectiveness AAAI 2025
Dynamic-Width Speculative Beam Decoding for LLM Inference AAAI 2025
Don’t Think It Twice: Exploit Shift Invariance for Efficient Online Streaming Inference of CNNs AAAI 2025
PLADIS: Pushing the Limits of Attention in Diffusion Models at Inference Time by Leveraging Sparsity ICCV 2025
Exploring Hybrid Sampling Inference for Aspect-based Sentiment Analysis NAACL 2025
Flaming-hot Initiation with Regular Execution Sampling for Large Language Models NAACL 2025
SwiftKV: Fast Prefill-Optimized Inference with Knowledge-Preserving Model Transformation EMNLP 2025
Unleashing Vecset Diffusion Model for Fast Shape Generation ICCV 2025
Cacheback: Speculative Decoding With Nothing But Cache EMNLP 2025