ICLR 2025 Conference Submissions

Information Bottleneck for Active Feature Acquisition
Alexander Luke Ian Norcliffe, Changhee Lee, Fergus Imrie, Mihaela van der Schaar, Pietro Lio
- 27 Sept 2024 (modified: 05 Feb 2025)
- Submitted to ICLR 2025
- Readers: Everyone
Q-SFT: Q-Learning for Language Models via Supervised Fine-Tuning
Joey Hong, Anca Dragan, Sergey Levine
- Published: 22 Jan 2025, Last Modified: 16 Mar 2025
- ICLR 2025 Poster
- Readers: Everyone
MM-EMBED: UNIVERSAL MULTIMODAL RETRIEVAL WITH MULTIMODAL LLMS
Sheng-Chieh Lin, Chankyu Lee, Mohammad Shoeybi, Jimmy Lin, Bryan Catanzaro, Wei Ping
- Published: 22 Jan 2025, Last Modified: 02 Mar 2025
- ICLR 2025 Poster
- Readers: Everyone
Spark Transformer: How Many FLOPs is a Token Worth?
Chong You, Kan Wu, Zhipeng Jia, Lin Chen, Srinadh Bhojanapalli, Praneeth Netrapalli, Jeremiah J. Willcock, Suvinay Subramanian, Felix Chern, Alek Andreev, Shreya Pathak, Felix Yu, Prateek Jain, David E Culler, Henry Levy, Sanjiv Kumar
- 27 Sept 2024 (modified: 05 Feb 2025)
- Submitted to ICLR 2025
- Readers: Everyone
QuantRad: Advancing Quantitative Reliability in Radiology Report Generation with Cascaded Decoders
Ying Jin, Noel C Codella, Yanbo Xu, Haoquan Fang, Yu Gu, Mu Wei, Jenq-Neng Hwang
- 27 Sept 2024 (modified: 31 Oct 2024)
- ICLR 2025 Conference Withdrawn Submission
- Readers: Everyone
On more accurate alignment modeling methods for automatic speech recognition
Albert Zeyer, Tina Raissi, Ralf Schlüter, Hermann Ney
- 27 Sept 2024 (modified: 03 Dec 2024)
- ICLR 2025 Conference Withdrawn Submission
- Readers: Everyone
In Praise of Stubbornness: The Case for Cognitive-Dissonance Aware Continual Update of Knowledge in LLMs
Simone Clemente, Zied Ben Houidi, Alexis Huet, Dario Rossi, Giulio Franzese, Pietro Michiardi
- 27 Sept 2024 (modified: 05 Feb 2025)
- Submitted to ICLR 2025
- Readers: Everyone
RNNs are not Transformers (Yet): The Key Bottleneck on In-Context Retrieval
Kaiyue Wen, Xingyu Dang, Kaifeng Lyu
- Published: 22 Jan 2025, Last Modified: 02 Mar 2025
- ICLR 2025 Poster
- Readers: Everyone
DC-Spin: A Speaker-invariant Speech Tokenizer For Spoken Language Models
Heng-Jui Chang, Hongyu Gong, Changhan Wang, James R. Glass, Yu-An Chung
- 27 Sept 2024 (modified: 05 Feb 2025)
- Submitted to ICLR 2025
- Readers: Everyone
Interactive Dialogue Agents via Reinforcement Learning with Hindsight Regenerations
Joey Hong, Jessica Lin, Sergey Levine, Anca Dragan
- 27 Sept 2024 (modified: 05 Feb 2025)
- Submitted to ICLR 2025
- Readers: Everyone
Regressing the Relative Future: Efficient Policy Optimization for Multi-turn RLHF
Zhaolin Gao, Wenhao Zhan, Jonathan Daniel Chang, Gokul Swamy, Kianté Brantley, Jason D. Lee, Wen Sun
- Published: 22 Jan 2025, Last Modified: 28 Feb 2025
- ICLR 2025 Poster
- Readers: Everyone
Efficiently Identifying Watermarked Segments in Mixed-Source Texts
Xuandong Zhao, Chenwen Liao, Yu-Xiang Wang, Lei Li
- 27 Sept 2024 (modified: 07 Dec 2024)
- ICLR 2025 Conference Withdrawn Submission
- Readers: Everyone
Foundation Vision Models are Unsupervised Image Canonicalizers
Utkarsh Singhal, Ryan Feng, Stella X. Yu, Atul Prakash
- 27 Sept 2024 (modified: 23 Jan 2025)
- ICLR 2025 Conference Withdrawn Submission
- Readers: Everyone
Diffusing States and Matching Scores: A New Framework for Imitation Learning
Runzhe Wu, Yiding Chen, Gokul Swamy, Kianté Brantley, Wen Sun
- Published: 22 Jan 2025, Last Modified: 01 Mar 2025
- ICLR 2025 Poster
- Readers: Everyone
Linear Representations of Political Perspective Emerge in Large Language Models
Junsol Kim, James Evans, Aaron Schein
- Published: 22 Jan 2025, Last Modified: 02 Apr 2025
- ICLR 2025 Oral
- Readers: Everyone
Understanding the Interplay between Parametric and Contextual Knowledge for Large Language Models
Sitao Cheng, Liangming Pan, Xunjian Yin, Xinyi Wang, William Yang Wang
- 27 Sept 2024 (modified: 05 Feb 2025)
- Submitted to ICLR 2025
- Readers: Everyone
Towards continuous machine learning on periodic crystals by ultra-fast invariants
Daniel Eric Widdowson, Vitaliy Kurlin
- 27 Sept 2024 (modified: 05 Feb 2025)
- Submitted to ICLR 2025
- Readers: Everyone
Behavioral Entropy-Guided Dataset Generation for Offline Reinforcement Learning
Wesley A. Suttle, Aamodh Suresh, Carlos Nieto-Granda
- Published: 22 Jan 2025, Last Modified: 27 Feb 2025
- ICLR 2025 Poster
- Readers: Everyone
Resolving Domain Shift For Representations Of Speech In Non-Invasive Brain Recordings
Jeremiah Ridge, Oiwi Parker Jones
- 27 Sept 2024 (modified: 05 Feb 2025)
- Submitted to ICLR 2025
- Readers: Everyone
Small features matter: Robust representation for world models
Zarif Ikram, Miranda Anna Christ, Ling Pan, Dianbo Liu
- 27 Sept 2024 (modified: 05 Feb 2025)
- Submitted to ICLR 2025
- Readers: Everyone
Rational Metareasoning for Large Language Models
C. Nicolò De Sabbata, Theodore Sumers, Thomas L. Griffiths
- 27 Sept 2024 (modified: 05 Feb 2025)
- Submitted to ICLR 2025
- Readers: Everyone
Context-aware Dynamic Pruning for Speech Foundation Models
Masao Someki, Yifan Peng, Siddhant Arora, Markus Müller, Athanasios Mouchtaris, Grant Strimel, Jing Liu, Shinji Watanabe
- Published: 22 Jan 2025, Last Modified: 15 Mar 2025
- ICLR 2025 Poster
- Readers: Everyone
Consistency-based Black-box Uncertainty Quantification for Text-to-SQL by Similarity Aggregation
Debarun Bhattacharjya, Balaji Ganesan, Radu Marinescu, Katsiaryna Mirylenka, Junkyu Lee, Michael Glass, Xiao Shou
- 27 Sept 2024 (modified: 15 Dec 2024)
- ICLR 2025 Conference Withdrawn Submission
- Readers: Everyone
From Pixels to Prose: A Large Dataset of Dense Image Captions
Vasu Singla, Kaiyu Yue, Sukriti Paul, Reza Shirkavand, Bethmage Mayuka Jayawardhana, Alireza Ganjdanesh, Heng Huang, Abhinav Bhatele, Gowthami Somepalli, Tom Goldstein
- 27 Sept 2024 (modified: 09 Dec 2024)
- ICLR 2025 Conference Withdrawn Submission
- Readers: Everyone
Infinite-Resolution Integral Noise Warping for Diffusion Models
Yitong Deng, Winnie Lin, Lingxiao Li, Dmitriy Smirnov, Ryan D Burgert, Ning Yu, Vincent Dedun, Mohammad H. Taghavi
- Published: 22 Jan 2025, Last Modified: 14 Mar 2025
- ICLR 2025 Poster
- Readers: Everyone