How DeepSeek’s DSpark Makes LLMs 85% Faster Without Retraining

DeepSeek has introduced DSpark, an open-source speculative decoding method that accelerates LLM inference by up to 85% ...
techtimes

NVIDIA’s New LLM Decodes 6x More Tokens Without an Auxiliary Draft Model

Any engineer currently running Eagle3 or a multi-token-prediction setup to speed up their LLM inference now has a compelling reason to look at the alternative NVIDIA's research team published Tuesday.

DeepSeek open sources DSpark, a new framework to speed up LLM inference by up to 85%

DSpark can make decoding faster, but acceptance quality still determines how much speed the system actually realizes.
GitHub

path_attention_position_encoding_via_accumulating_householder_transformations.md

description [NeurIPS 2025][LLM Evaluation][position encoding] This paper proposes PaTH (Position encoding via accumulating Householder Transformations), a data-dependent multiplicative position ...
Frontiers

Augmenting large language models with psychologically grounded models of causal reasoning for planning under uncertainty

Large Language Models (LLMs) have come a long way in their ability to solve a wide range of problems. Yet, LLM decision-making still relies primarily on pattern recognition, which may limit its ...
GitHub

RF-Agent: Automated Reward Function Design via Language Agent Tree Search

Designing efficient reward functions for low-level control tasks is a challenging problem. While recent research uses Large Language Models (LLMs) to generate dense reward functions, they often suffer ...
Frontiers

The art of audience engagement: LLM-based thin-slicing of scientific talks

This paper examines the thin-slicing approach—the ability to make accurate judgments based on minimal information—in the context of scientific presentations. Drawing on research from nonverbal ...