Hiroki Furuta
13 papers · 2021–2026 · 4 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+5 more ↓ Show less ↑
πΊοΈ Taxonomy Completionist (15) π§ Keyword Pioneer π Conference Polyglot (3) π Cross-Pollinator (13) π Renaissance Researcher (5)
π
Interdisciplinary Bridge
π
Triple Crown
π€
Dynamic Duo
(10)
π₯
Unstoppable
(5)
π
Century Club
(12)
Conferences
ICLR (6)
ICML (4)
NIPS (2)
ACL (1)
Top co-authors
Research topics
Keywords
deep reinforcement learning
(2)
reinforcement learning
(1)
preference learning
(1)
direct preference optimization
(1)
language model alignment
(1)
reinforcement learning from human feedback
(1)
model alignment
(1)
mutual information
(1)
loss function
(1)
reward model
(1)
reward shaping
(1)
mechanistic interpretability
(1)
sparse autoencoder
(1)
soft label
(1)
geometric average
(1)
information-theoretic measure
(1)
preference distribution
(1)
large language model alignment
(1)
task difficulty
(1)
task complexity
(1)
Papers
Understanding Emergent Misalignment via Feature Superposition Geometry
ACL 2026
Plan-and-Act: Improving Planning of Agents for Long-Horizon Tasks
ICML 2025
Rethinking Evaluation of Sparse Autoencoders through the Representation of Polysemous Words
ICLR 2025
Beyond Induction Heads: In-Context Meta Learning Induces Multi-Phase Circuit Emergence
ICML 2025
A Real-World WebAgent with Planning, Long Context Understanding, and Program Synthesis
ICLR 2024
Geometric-Averaged Preference Optimization for Soft Preference Labels
NIPS 2024
Multimodal Web Navigation with Instruction-Finetuned Foundation Models
ICLR 2024
A Human-Inspired Reading Agent with Gist Memory of Very Long Contexts
ICML 2024
A System for Morphology-Task Generalization via Unified Representation and Behavior Distillation
ICLR 2023
Generalized Decision Transformer for Offline Hindsight Information Matching
ICLR 2022
Policy Information Capacity: Information-Theoretic Measure for Task Complexity in Deep Reinforcement Learning
ICML 2021
Deployment-Efficient Reinforcement Learning via Model-Based Offline Optimization
ICLR 2021
Co-Adaptation of Algorithmic and Implementational Innovations in Inference-based Deep Reinforcement Learning
NIPS 2021