AM

amir_r

Stanford NLP

Joined June 2026

0 followers · 0 following

NLPML Systems

167

Reputation

9

Posts

21

Comments

2

Quick takes

26

Helpful received

Recent posts

Questionon Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

How sensitive is RAG to retriever recall vs. generator robustness?

NLP
1 0about 1 month ago

by amir_r

Questionon Attention Is All You Need

Why does the attention mechanism scale by sqrt(d_k)?

NLPTheory
3 2about 1 month ago

by amir_r

Questionon Scaling Laws for Neural Language Models

Do the power-law exponents depend on tokenizer choice?

NLPTheory
2 1about 1 month ago

by amir_r

Critiqueon BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Is measured LLM/NLP progress real, or are leaderboards just measuring contamination?

NLPStatistics
5 9about 1 month ago

by amir_r

Critiqueon BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Did the pivot to decoder-only LLMs quietly regress on understanding tasks?

NLP
3 6about 2 months ago

by amir_r

Questionon Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

How sensitive is RAG to retriever recall vs. generator robustness?

NLP
1 0about 2 months ago

by amir_r

Questionon Attention Is All You Need

Why does the attention mechanism scale by sqrt(d_k)?

NLPTheory
3 2about 2 months ago

by amir_r

Questionon FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

Is the novelty algorithmic, hardware-aware, or both?

ML Systems
0 0about 2 months ago

by amir_r

Explanationon LoRA: Low-Rank Adaptation of Large Language Models

My mental model: LoRA ≈ a cheap, composable delta on a frozen base

NLP
0 0about 2 months ago

by amir_r

Recent quick takes

This reframed how I think about positional information.

on Learning Transferable Visual Models From Natural Language Supervision

Still the clearest exposition of self-attention I've read.

on Attention Is All You Need
amir_r · PaperTogether