16 papers
Graph Coarsening Enables Escaping Local Optima in Compute-Constrained Nonconvex MAX-CUT
M. Harper Langston, Pierre-David Letourneau, Dalton Jones +7
Because its binary optimization landscape contains exponentially many local optima, MAX-CUT is highly sensitive to initialization. We study Particle Polynomial Relaxation (PPR), a…
ValueDiff: Value-Geometric KV Cache Eviction for Sink-Suppressed LLMs
Junyoung Park, Jungwook Choi, Mingu Lee
Modern LLMs with QK-normalization, gated attention, learned attention sinks, or logit softcapping exhibit weaker persistent attention sinks, on which existing KV cache eviction met…
Efficient Training-Free Multi-Token Prediction via Embedding-Space Probing
Raghavv Goel, Mukul Gagrani, Mingu Lee +1
Large Language Models (LLMs) possess latent multi-token prediction (MTP) abilities despite being trained only for next-token generation. We introduce ESP (Embedding-Space Probing),…
QUOKA: Query-Oriented KV Selection For Efficient LLM Prefill
Dalton Jones, Junyoung Park, Matthew Morse +3
We present QUOKA: Query-oriented KV selection for efficient attention, a training-free and hardware agnostic sparse attention algorithm for accelerating transformer inference under…
Double-P: Hierarchical Top-P Sparse Attention for Long-Context LLMs
Wentao Ni, Kangqi Zhang, Zhongming Yu +7
As long-context inference becomes central to large language models (LLMs), attention over growing key-value caches emerges as a dominant decoding bottleneck, motivating sparse atte…
Fast Forward: Accelerating LLM Prefill with Predictive FFN Sparsity
Aayush Gautam, Mukul Gagrani, Junyoung Park +3
The prefill stage of large language model (LLM) inference is a key computational bottleneck for long-context workloads. At short-to-moderate context lengths (1K--16K tokens), Feed-…
Structuring The Future: Diffusion LLM Speculative Decoding via Calibrated Draft Graphs
Sudhanshu Agrawal, Risheek Garrepalli, Raghavv Goel +3
Diffusion LLMs (dLLMs) have recently emerged as a powerful alternative to autoregressive LLMs (AR-LLMs) with the potential to operate at significantly higher token-generation rates…
VOCABTRIM: Vocabulary Pruning for Efficient Speculative Decoding in LLMs
Raghavv Goel, Sudhanshu Agrawal, Mukul Gagrani +9
In this paper, we introduce a simple training-free technique to improve the performance of drafter-based speculative decoding (SpD) methods that incorporates language modeling head…
KeyDiff: Key Similarity-Based KV Cache Eviction for Long-Context LLM Inference in Resource-Constrained Environments
Junyoung Park, Dalton Jones, Matthew J Morse +3
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 evi…
CAOTE: KV Cache Selection for LLMs via Attention Output Error-Based Token Eviction
Raghavv Goel, Junyoung Park, Mukul Gagrani +5
While long context support of large language models has extended their abilities, it also incurs challenges in memory and compute which becomes crucial bottlenecks in resource-rest…
BreezyVoice: Adapting TTS for Taiwanese Mandarin with Enhanced Polyphone Disambiguation -- Challenges and Insights
Chan-Jan Hsu, Yi-Cheng Lin, Chia-Chun Lin +10
We present BreezyVoice, a Text-to-Speech (TTS) system specifically adapted for Taiwanese Mandarin, highlighting phonetic control abilities to address the unique challenges of polyp…
AdaEDL: Early Draft Stopping for Speculative Decoding of Large Language Models via an Entropy-based Lower Bound on Token Acceptance Probability
Sudhanshu Agrawal, Wonseok Jeon, Mingu Lee
Speculative decoding is a powerful technique that attempts to circumvent the autoregressive constraint of modern Large Language Models (LLMs). The aim of speculative decoding techn…
ToSA: Token Selective Attention for Efficient Vision Transformers
Manish Kumar Singh, Rajeev Yasarla, Hong Cai +2
In this paper, we propose a novel token selective attention approach, ToSA, which can identify tokens that need to be attended as well as those that can skip a transformer layer. M…
On Speculative Decoding for Multimodal Large Language Models
Mukul Gagrani, Raghavv Goel, Wonseok Jeon +3
Inference with Multimodal Large Language Models (MLLMs) is slow due to their large-language-model backbone which suffers from memory bandwidth bottleneck and generates tokens auto-…
Recursive Speculative Decoding: Accelerating LLM Inference via Sampling Without Replacement
Wonseok Jeon, Mukul Gagrani, Raghavv Goel +3
Speculative decoding is an inference-acceleration method for large language models (LLMs) where a small language model generates a draft-token sequence which is further verified by…
Direct Alignment of Draft Model for Speculative Decoding with Chat-Fine-Tuned LLMs
Raghavv Goel, Mukul Gagrani, Wonseok Jeon +3
Text generation with Large Language Models (LLMs) is known to be memory bound due to the combination of their auto-regressive nature, huge parameter counts, and limited memory band…