DAIR.AI · Curated weekly since April 2023
AI Papers of the Week
Every paper worth reading in AI, hand-picked one week at a time.
1,760
Papers
176
Weekly issues
2023
Since
173 papers · Reinforcement LearningClear filters →
SequenceMatch
Formulates sequence generation as imitation learning, enabling backtracking via a backspace action.
AlphaDev
DeepMind's deep RL agent discovering faster sorting algorithms from scratch, now in LLVM.
Fine-Grained RLHF
Trains LMs with segment-level human feedback rather than whole-response preferences.
Let's Verify Step by Step
OpenAI's landmark paper on process reward models for mathematical reasoning.
Direct Preference Optimization (DPO)
Rafailov et al.'s simpler alternative to RLHF that rivals full RL-based alignment.
176 weeks of AI research · papers per week
Week of Aug 17–23, 202610 papers →
Apr2023
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan2024
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan2025
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jan2026
Feb
Mar
Apr
May
Jun
Jul
Aug
Apr 2023Hover a week to inspect · select to openAug 2026