HNHacker News
TopNewBestAskShowJobs

veryluckyxyz

549 karma · joined September 30, 2014

submissionscomments

Scaling Laws for Agent Harnesses via Effective Feedback Compute

arxiv.org·1 pts·veryluckyxyz·
0

Generalizing Test-Time Compute-Optimal Scaling as an Optimizable Graph

huggingface.co·2 pts·veryluckyxyz·
0

Hidden drivers of HRM's performance on ARC-AGI

arcprize.org·31 pts·veryluckyxyz·
2

Set Block Decoding Is a Language Model Inference Accelerator

arxiv.org·4 pts·veryluckyxyz·
0

Deep Think with Confidence

jiaweizzhao.github.io·1 pts·veryluckyxyz·
0

A Batch Size and Token NUM- BER Agnostic Learning Rate Scheduler

arxiv.org·2 pts·veryluckyxyz·
0

Easily Understand Rdma Technology

naddod.com·1 pts·veryluckyxyz·
1

Model Merging in Pre-Training of Large Language Models

arxiv.org·2 pts·veryluckyxyz·
0

Understanding Perception and Reasoning Through Model Merging

arxiv.org·2 pts·veryluckyxyz·
0

Building and better understanding vision-language models (2024)

huggingface.co·2 pts·veryluckyxyz·
0

HF smolagents computer-agent demo

huggingface.co·1 pts·veryluckyxyz·
0

Do Reasoning Models Show Better Verbalized Calibration?

arxiv.org·2 pts·veryluckyxyz·
0

Robustly identifying concepts introduced during chat fine-tuning with crosscoder

arxiv.org·6 pts·veryluckyxyz·
0

Retrieval with Learned Similarities

arxiv.org·3 pts·veryluckyxyz·
0

The Curse of Depth in Large Language Models

arxiv.org·1 pts·veryluckyxyz·
0

Looking Back at Speculative Decoding

research.google·36 pts·veryluckyxyz·
5

Long-Context GRPO

unsloth.ai·60 pts·veryluckyxyz·
22

HippoRAG: Neurobiologically Inspired Long-Term Memory for LLMs (2024)

arxiv.org·65 pts·veryluckyxyz·
4

Learning to Plan and Reason for Evaluation with Thinking-LLM-as-a-Judge

arxiv.org·1 pts·veryluckyxyz·
0

Process Reinforcement Through Implicit Rewards

curvy-check-498.notion.site·1 pts·veryluckyxyz·
0

Explaining Large Language Models Decisions Using Shapley Values

arxiv.org·89 pts·veryluckyxyz·
19

Phi-4 Technical Report

arxiv.org·2 pts·veryluckyxyz·
0

Alignment Faking in LLMs [pdf]

assets.anthropic.com·2 pts·veryluckyxyz·
1

What Makes Rotary Positional Encodings Useful?

arxiv.org·1 pts·veryluckyxyz·
0

Rethinking Softmax: Self-Attention with Polynomial Activations

arxiv.org·2 pts·veryluckyxyz·
0

Post-Training Layer Scaling Prevents Forgetting and Enhances Model Merging

arxiv.org·1 pts·veryluckyxyz·
0

Random Matrix Theory in Machine Learning Tutorial

random-matrix-learning.github.io·2 pts·veryluckyxyz·
0

Rerankers: A Lightweight Python Library to Unify Ranking Methods

answer.ai·1 pts·veryluckyxyz·
0

Double Descent Demystified

arxiv.org·1 pts·veryluckyxyz·
0

Synthetic Continued Pretraining

arxiv.org·3 pts·veryluckyxyz·
0
Page 1 of 2Next →