@manasvardhan
I write philosophy, literature and tech
Nothing here yet.
Nothing here yet.
5d ago · 7 min read · RLVR (Reinforcement Learning with Verifiable Rewards) has been the go-to recipe for boosting LLM reasoning since DeepSeek-R1 made it mainstream. The formula is simple: give the model math problems, check the answers, reward correct reasoning chains. ...
Join discussionApr 4 · 8 min read · Everyone's building AI research agents. Feed them a Kaggle problem, let them explore, iterate, and submit. The promise: autonomous AI that does your ML engineering while you sleep. The reality: most of these systems plateau around 65-70% on benchmark...
Join discussionApr 2 · 6 min read · Here's the deal we thought we had with chain-of-thought prompting: let the model show its work, and we can watch the reasoning unfold. If something goes wrong, we'd see it in the chain. CoT was our audit trail, our interpretability shortcut, our free...
Join discussionApr 2 · 7 min read · For the last two years, the LLM inference community has been playing a game of architectural bingo. Multi-Head Attention (MHA)? Too expensive at scale. Grouped-Query Attention (GQA)? Better KV cache, but you lose expressiveness. Multi-Head Latent Att...
Join discussionMar 29 · 7 min read · The most dangerous failure mode in production LLMs isn't hallucination. It's confident hallucination. The model is dead wrong, and it's absolutely certain about it. Token entropy is low. Confidence scores look great. Your monitoring dashboard shows g...
MAamer commented