EuroSafeGuard v1
Collection
Open constitutional classifiers: a probe-escalated guardrail for LLMs, its CC++ recreation and training prompts. • 3 items • Updated • 3
🔄 In a Training Loop
Pratyay Banerjee
Neilblaze
AI & ML interests
IR, NLP, Pattern Recognition, xAI, Interpretability, Evals
Recent Activity
upvoted a collection 5 days ago
EuroSafeGuard v1 liked a model 5 days ago
Qwen/Qwen3Guard-Gen-0.6B liked a model 5 days ago
Qwen/Qwen3Guard-Stream-0.6BOrganizations
upvoted a collection 5 days ago
liked 2
models 5 days ago
upvoted a collection 5 days ago
liked 2
models 5 days ago
upvoted a paper 5 days ago
liked 7
models 5 days ago
nace-ai/drex-v2.0
35B • Updated • 469 • 23
HAL-X-AI/THX-02
Zero-Shot Classification • 5B • Updated • 2.13k • 50
orcarouter/LFM3-1.7B-abliterated
Text Generation • 2B • Updated • 681 • 45
Jiunsong/SuperDeepseek-V4-Flash-abliterated-MQ-2xDGX
Text Generation • 306B • Updated • 518 • 64
moonshotai/Kimi-K3.1-Pro
Image-Text-to-Text • 2.8T • Updated • 33.6k • • 1.93k
ricursiveAI/Ripple-65B-Preview
65B • Updated • 7.11k • 515
nace-ai/drex-max-v1
Zero-Shot Classification • 26B • Updated • 567 • 46
liked a
model 6 days ago
upvoted 5
papers 8 days ago
What Effort Level Should I Choose? On the Economics of Reasoning Effort in LLM Agents
Paper • 2610.34491 • Published • 22
Channel Selection for Mixture-of-Experts Decoding Needs No Calibration Data
Paper • 2610.37386 • Published • 26
S3AC: Stabilized, Selective, and Scalable Actor-Critic for LLM Reasoning
Paper • 2610.35518 • Published • 18
WebAnything: From the Wild Web to Scalable Agent Environments
Paper • 2610.37786 • Published • 19
Memory Tax: Dissecting the Net Value of Agent Memory
Paper • 2610.34073 • Published • 27