Article
Prompt Phrasing Shifts Model Performance More Than Tier — and Which LLMs Will Get Your Jokes
Wayfinder6
• • 2
🔄 In a Training Loop
Pratyay Banerjee
Neilblaze
AI & ML interests
IR, NLP, Pattern Recognition, xAI, Interpretability, Evals
Recent Activity
upvoted an article about 3 hours ago
Prompt Phrasing Shifts Model Performance More Than Tier — and Which LLMs Will Get Your Jokes liked a model 1 day ago
LiquidAI/LFM2.5-230M liked a model 5 days ago
unsloth/GLM-5.2-GGUFOrganizations
upvoted an article about 3 hours ago
upvoted 5
papers 8 days ago
Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents
Paper • 2606.06036 • Published • 75
FastContext: Training Efficient Repository Explorer for Coding Agents
Paper • 2606.14066 • Published • 93
Toward Generalist Autonomous Research via Hypothesis-Tree Refinement
Paper • 2606.11926 • Published • 121
EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments
Paper • 2606.13681 • Published • 142
Looped World Models
Paper • 2606.18208 • Published • 468
upvoted 6
papers 11 days ago
Measuring Epistemic Resilience of LLMs Under Misleading Medical Context
Paper • 2606.12291 • Published • 60
Learning from the Self-future: On-policy Self-distillation for dLLMs
Paper • 2606.18195 • Published • 76
Robust-U1: Can MLLMs Self-Recover Corrupted Visual Content for Robust Understanding?
Paper • 2606.08063 • Published • 81
Redesign Mixture-of-Experts Routers with Manifold Power Iteration
Paper • 2606.12397 • Published • 89
VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Models
Paper • 2606.16140 • Published • 119
Your UnEmbedding Matrix is Secretly a Feature Lens for Text Embeddings
Paper • 2606.07502 • Published • 99
upvoted an article 17 days ago
Article
Build Small Hackathon With Cohere Models
CohereLabs
• • 5 upvoted 7
papers 18 days ago
OpenSkill: Open-World Self-Evolution for LLM Agents
Paper • 2606.06741 • Published • 29
Rethinking the Divergence Regularization in LLM RL
Paper • 2606.09821 • Published • 33
SearchSwarm: Towards Delegation Intelligence in Agentic LLMs for Long-Horizon Deep Research
Paper • 2606.09730 • Published • 54
Retrospective Harness Optimization: Improving LLM Agents via Self-Preference over Trajectory Rollouts
Paper • 2606.05922 • Published • 69
LatentSkill: From In-Context Textual Skills to In-Weight Latent Skills for LLM Agents
Paper • 2606.06087 • Published • 66
KVarN: Variance-Normalized KV-Cache Quantization Mitigates Error Accumulation in Reasoning Tasks
Paper • 2606.03458 • Published • 67
Agents' Last Exam
Paper • 2606.05405 • Published • 368