Mutual Information

Feeds to Scour
SubscribedAll
Scoured 72 posts in 7.2 ms

KaiFelixBennett/gemma4-turboquant-rdna4: Run Gemma-4-31B at full 256K context on a $1,400 AMD RDNA4 GPU (gfx1201): TurboQuant KV cache + HIP-graph-safe Flash-Attention for llama.cpp, fully measured on real hardware.

聽馃Local LLMs 聽Content type: Code
github.com路路Hacker News

A machine-learning-assisted progressive digit-randomness screening framework for detecting non-random patterns in raw numerical research data

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

Optimality of FSQ Tokens for Continuous Diffusion for Categorical Data with Application to Text-to-Speech

聽馃挰Natural Language Processing 聽Content type: Academic
arxiv.org路

Mutual Information Optimization via K-Recursion and Automatic Differentiation for Linear Gaussian Wireless Networks

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

Flow-DPPO: Divergence Proximal Policy Optimization for Flow Matching Models

聽馃敘Kolmogorov Complexity 聽Content type: Academic
arxiv.org路

SynIB: Informational Bottleneck for Maximizing Synergy in Multimodal Learning

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

Optimal Feedback Communication with Information Maximization and Distortion Minimization

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

Prime Event Languages: An Information-Theoretic Investigation of Twin-Prime Event Structure

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

Self-Distilled Policy Gradient

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

Support sufficiency as action-sufficient compression: a single-cycle rate-regret formulation

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

Causally Evaluating the Learnability of Formal Language Tasks

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

Towards Pretraining Text Encoders for TabPFN

聽馃ЛContent Discovery 聽Content type: Academic
arxiv.org路

Style or Content? Evaluating Style Classifiers with Controlled Content Overlap

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

boostsecurityio/bagel: bagel, a CLI that inventories security-relevant metadata on developer workstations

聽鈩癸笍Information Theory 聽Content type: Code
github.com路

Minimum Distortion Quantization with Specified Output Distribution

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

FAIR-Calib: Frontier-Aware Instability-Reweighted Calibration for Post-Training Quantization of Diffusion Large Language Models

聽馃敘Kolmogorov Complexity 聽Content type: Academic
arxiv.org路

Beyond Output Matching: Preserving Internal Geometry in NVFP4 LLM Distillatio

聽馃敆RAG 聽Content type: Academic
arxiv.org路

Distilling Safe LLM Systems via Soft Prompts for On Device Settings

聽馃Local LLMs 聽Content type: Academic
arxiv.org路

Robust Active Learning for Few-Shot Example Selection in Text-to-SQL

聽馃МAlgorithms 聽Content type: Academic
arxiv.org路

Improving the Efficiency and Effectiveness of LLM Knowledge Distillation for Conversational Search

聽鈩癸笍Information Theory 聽Content type: Academic
arxiv.org路

Keyboard Shortcuts

Navigation

Next / previous item
j/k
Open post
oorEnter
Preview post
v

Post Actions

Love post
a
Like post
l
Dislike post
d
Undo reaction
u
Save / unsave
s

Recommendations

Add interest / feed
Enter
Not interested
x

Go to

Home
gh
Interests
gi
Feeds
gf
Likes
gl
History
gy
Changelog
gc
Settings
gs
Browse
gb
Search
/

General

Show this help
?
Submit feedback
!
Close modal / unfocus
Esc

Press ? anytime to show this help