ZK
Hamidun News Editorial Desk · Editor-in-Chief Zhemal Khamidun

Zhemal Khamidun

Hamidun News editorial desk led by Editor-in-Chief Zhemal Khamidun (CPO of AlpinaGPT). Daily curation across 50+ global AI labs, research feeds, and industry reports in 6 languages.

10,282Published articles
50+Verified primary sources
6Language editions (RU/EN/ES/PT/FR/AR)
D-SLR Replaces Truncated SVD With Three Decompositions and No Tuning
Research·arXiv cs.LG

D-SLR Replaces Truncated SVD With Three Decompositions and No Tuning

Oct 9, 2026·2 min
Phase-HDC trains a model with no optimizer history and stores 16–23x less than Adam
Hardware

Phase-HDC trains a model with no optimizer history and stores 16–23x less than Adam

Oct 9, 2026·3 min
Kolmogorov-Arnold Networks Match MLP Performance With 40% Fewer Samples
Robotics

Kolmogorov-Arnold Networks Match MLP Performance With 40% Fewer Samples

Oct 9, 2026·2 min
IKNO for Navier–Stokes Stays Accurate Under Rotation, With 5.6x Fewer Parameters Than TFN
Research

IKNO for Navier–Stokes Stays Accurate Under Rotation, With 5.6x Fewer Parameters Than TFN

Oct 9, 2026·2 min
LoopOPD Teaches Looped Language Models to Improve Through Their Own Loops
Research·arXiv cs.LG

LoopOPD Teaches Looped Language Models to Improve Through Their Own Loops

LoopOPD makes an intermediate loop of a LoopLM learn from the final loop of the same model, with no external teacher. Tested on Ouro-Thinking, trained on math only.

Oct 9, 2026·2 min
MoE Router Telemetry Leaks Fine-Tuning Data: Attack Gains 2.7–9.4 pp in Accuracy
Research·arXiv cs.LG

MoE Router Telemetry Leaks Fine-Tuning Data: Attack Gains 2.7–9.4 pp in Accuracy

MoE expert-routing logs reveal whether an example was in the fine-tuning set: TPR at 1% FPR rises by 2.7–9.4 pp in all nine tested configurations.

Oct 9, 2026·3 min
SPERA Trained on 80,000 Hours of EEG From 29,048 People and Tested on 9 Tasks
Medicine·arXiv cs.LG

SPERA Trained on 80,000 Hours of EEG From 29,048 People and Tested on 9 Tasks

The SPERA foundation model predicts EEG in latent space using a JEPA scheme and accounts for electrode geometry. Training data: 106 datasets, about 80,000 hours of…

Oct 9, 2026·2 min
80 Synthetic Dialogues Are Enough to Train Activation Probes for Gemma-3-27B, but Not Everywhere
Research·arXiv cs.LG

80 Synthetic Dialogues Are Enough to Train Activation Probes for Gemma-3-27B, but Not Everywhere

80 synthetic examples are enough to monitor "high-stakes" situations and harmful responses. Instruction-following probes need several times more. It all comes down…

Oct 9, 2026·2 min
Top themes
01Temporal Transformer for the CAN Bus Catches Anomalies Without Sharing Raw DataOct 9, 2026·2 min02Freezing the Decoder in KV-Cache Compression Saves 3x Memory but Doesn't Improve AccuracyOct 9, 2026·2 min03SFT-as-Context Restores General Skills in Fine-Tuned LLMs, Within 2.2 PointsOct 9, 2026·3 min04Lapras replaces CoT with latent reasoning: up to +10.79% F1 with 23.9x fewer tokensOct 9, 2026·2 min05Doctors Don't Ask LLMs About Diagnoses: 127 833 Queries vs. 58 BenchmarksOct 9, 2026·2 min
FedAlphaEdit Merges Knowledge Edits from Different Organizations into One LLM Without Sharing Data
Research·arXiv cs.CL

FedAlphaEdit Merges Knowledge Edits from Different Organizations into One LLM Without Sharing Data

Oct 9, 2026·2 min
Three LLMs Follow Prompts for Stutters and Interruptions Poorly; Rules Do Better
Research·arXiv cs.CL

Three LLMs Follow Prompts for Stutters and Interruptions Poorly; Rules Do Better

Oct 9, 2026·2 min
Sociolinguistics Instead of LLM Judges: Testing User-Simulator Personas on 5 Agents
Research·arXiv cs.CL

Sociolinguistics Instead of LLM Judges: Testing User-Simulator Personas on 5 Agents

Oct 9, 2026·2 min
AI4Fire: SQL Tool Lifts LLM Accuracy on Wildfire Databases From 16% to 88%
Research·arXiv cs.CL

AI4Fire: SQL Tool Lifts LLM Accuracy on Wildfire Databases From 16% to 88%

Oct 9, 2026·2 min
LLMs Judge Translation Across 70 Language Pairs: They Beat Humans Only in Places
Research·arXiv cs.CL

LLMs Judge Translation Across 70 Language Pairs: They Beat Humans Only in Places

A study tested LLMs as machine translation judges under MQM and ESA across 70 language pairs and WMT23/WMT25 data. In places, models agree with humans better than…

Oct 9, 2026·2 min
STI-OPD: Teacher Probabilistically Swaps in Its Own Actions and Beats OPD Baselines
LLM·arXiv cs.CL

STI-OPD: Teacher Probabilistically Swaps in Its Own Actions and Beats OPD Baselines

New STI-OPD distillation method fixes trajectory drift in LLM agents: the teacher steps in with a KL-divergence-based probability and beats the best OPD baseline on…

Oct 9, 2026·2 min
MASA Reframes LLM Sparse Attention as Matrix Approximation
Research·arXiv cs.CL

MASA Reframes LLM Sparse Attention as Matrix Approximation

A new preprint proposes selecting sparse-attention elements by their contribution to matrix-product error, not by weight magnitude. The method plugs into existing sparse frameworks.

Oct 9, 2026·2 min
Routed Sparse Autoencoders Separate Words From Intonation in Speech on WavLM and SPEAR
Research·arXiv cs.CL

Routed Sparse Autoencoders Separate Words From Intonation in Speech on WavLM and SPEAR

Oct 9, 2026·2 min
Galahad-kv stores KV cache on NVMe and loads a block 2.8–4.3x faster than recomputing it
Research·arXiv cs.CL

Galahad-kv stores KV cache on NVMe and loads a block 2.8–4.3x faster than recomputing it

Oct 9, 2026·2 min
Fine-Tuned 0.8B-Parameter Model Beats GPT-5.4 at Grammar Annotation
Business·arXiv cs.CL

Fine-Tuned 0.8B-Parameter Model Beats GPT-5.4 at Grammar Annotation

Oct 9, 2026·2 min
Utterance Normalization Cuts the Cost of Analyzing Millions of Calls: A Pipeline Built Entirely on Small Models
Business·arXiv cs.CL

Utterance Normalization Cuts the Cost of Analyzing Millions of Calls: A Pipeline Built Entirely on Small Models

Oct 9, 2026·2 min
Text autoencoder compresses web text to 2.24 bits per byte without losing meaning
Research·arXiv cs.CL

Text autoencoder compresses web text to 2.24 bits per byte without losing meaning

Oct 9, 2026·2 min
Cognitive Thermometers: Neural Networks Proposed as a Way to Measure the Complexity of Meaning
Research

Cognitive Thermometers: Neural Networks Proposed as a Way to Measure the Complexity of Meaning

Oct 9, 2026·2 min
LLMs Draft Road-Work Plans for WisDOT: 14B Barely Beats 7B
Business

LLMs Draft Road-Work Plans for WisDOT: 14B Barely Beats 7B

Oct 9, 2026·2 min
Wieszcz-XIX: A 3.1-Billion-Word Polish Corpus From Before 1918 Trains Models From Scratch
Research

Wieszcz-XIX: A 3.1-Billion-Word Polish Corpus From Before 1918 Trains Models From Scratch

Oct 9, 2026·2 min
Diffu-LoRA Lets the Adapter Decide Which Stable Diffusion Layers Need Rank
Research·arXiv cs.CL

Diffu-LoRA Lets the Adapter Decide Which Stable Diffusion Layers Need Rank

A new method for personalizing diffusion models allocates LoRA rank across layers using learnable gates and pruning. Tested on Stable Diffusion and DreamBooth sets.

Oct 9, 2026·2 min
GenUI-Harness Lifts a 4B Model From 9.33% to 58% Pass@3, Beating Claude Opus 5
LLM·arXiv cs.AI

GenUI-Harness Lifts a 4B Model From 9.33% to 58% Pass@3, Beating Claude Opus 5

The multi-agent GenUI-Harness generates interfaces instead of text replies. After RL, a 4B model went from 9.33% to 58.00% Pass@3, and dialogue shrank from 3.4 to…

Oct 9, 2026·2 min