Goblin
News
AI news by
promptgoblins.ai
|
News
About
News
About
Filtered by:
post-training-tradeoffs
Clear
Titles
Summaries
March
6
OpenAI Researcher Shares Lessons on Evals, Post-Training, and AI Progress
Research
1
Mar 30
6
OpenAI Researcher Shares Lessons on Evals, Post-Training, and AI Progress
Research
· 1 src · Mar 30
Discuss
6.82
Semantic Calibration in LLMs: Why Base Models Know What They Know
Research
1
Mar 25
6.82
Semantic Calibration in LLMs: Why Base Models Know What They Know
Research
· 1 src · Mar 25
Discuss
7
Google DeepMind Algorithm Achieves 10x RLHF Data Efficiency
Research
1
Mar 20
7
Google DeepMind Algorithm Achieves 10x RLHF Data Efficiency
Research
· 1 src · Mar 20
Discuss
6
Why AI Still Cannot Write Well, Despite Ingesting All Literature
Research
1
Mar 17
6
Why AI Still Cannot Write Well, Despite Ingesting All Literature
Research
· 1 src · Mar 17
Discuss
Last Week
8
Anthropic: Teaching Claude Why Fixes Agentic Misalignment
Research
1
1d ago
8
Anthropic: Teaching Claude Why Fixes Agentic Misalignment
Top
Research
· 1 src · 1d ago
Discuss
6
OpenAI's Goblin Problem: How Reward Systems Create Self-Reinforcing AI Behavioral Attractors
Safety
1
2d ago
6
OpenAI's Goblin Problem: How Reward Systems Create Self-Reinforcing AI Behavioral Attractors
Safety
· 1 src · 2d ago
Discuss
2 Weeks Ago
6
Wing VC: Post-Training on Open-Weights Is the Strategic Battleground for AI App Companies
Research
1
Apr 28
6
Wing VC: Post-Training on Open-Weights Is the Strategic Battleground for AI App Companies
Research
· 1 src · Apr 28
Discuss
3 Weeks Ago
6
Perplexity Publishes Two-Stage Training Pipeline for Web Search Agents
Research
1
Apr 23
6
Perplexity Publishes Two-Stage Training Pipeline for Web Search Agents
Research
· 1 src · Apr 23
Discuss
7
RLVR Weak Supervision: When LLMs Can and Cannot Generalize
Research
1
Apr 22
7
RLVR Weak Supervision: When LLMs Can and Cannot Generalize
Research
· 1 src · Apr 22
Discuss
7
AI2 Introduces BAR: Modular Post-Training via Branch-Adapt-Route
Research
1
Apr 21
7
AI2 Introduces BAR: Modular Post-Training via Branch-Adapt-Route
Research
· 1 src · Apr 21
Discuss
Last Month
7
Simple Self-Distillation Boosts LLM Code Generation by 13 Points Without RL or Verifiers
Research
1
Apr 6
7
Simple Self-Distillation Boosts LLM Code Generation by 13 Points Without RL or Verifiers
Research
· 1 src · Apr 6
Discuss
Filters
Signal
Title
Category
Sources
Posted
Discuss
