Ax Ian Su, Gaurav Purushothaman, Jey Narayan, Ruhika Goel, Kevin Zhu, Sunishchal Dev, Yash More, Maheep Chaudhary 2/17/2026

Broken Chains: The Cost of Incomplete Reasoning in LLMs

Framework analyzing reasoning modalities (code, natural language, hybrid) in LLMs under token constraints, evaluating performance tradeoffs for reasoning-specialized models.

Ax Hasi Hays 2/17/2026

Selective Synchronization Attention

Novel attention mechanism (SSA) replacing dot-product self-attention with Kuramoto model solution, reducing quadratic complexity and grounding in biological neural computation.

Ax Chang Liu, Yiran Zhao, Lawrence Liu, Yaoqi Ye, Csaba Szepesv\'ari, Lin F. Yang 2/17/2026

LACONIC: Length-Aware Constrained Reinforcement Learning for LLM

Reinforcement learning approach (LACONIC) for controlling LLM response length during training without fixed heuristic reward shaping, addressing inference latency and computational overhead.

Ax Buze Zhang, Jinkai Tao, Zilang Zeng, Neil He, Ali Maatouk, Menglin Yang, Rex Ying 2/17/2026

Parameter-Efficient Fine-Tuning of LLMs with Mixture of Space Experts

Novel parameter-efficient fine-tuning method for LLMs using mixture of experts in alternative geometric spaces (hyperbolic, spherical) to capture complex language data structures.

Ax Shishir Sharma, Doina Precup, Theodore J. Perkins 2/17/2026

Fluid-Agent Reinforcement Learning

Framework for multi-agent RL with dynamic agent creation and reproduction, extending MARL beyond fixed agent counts.

Ax Karim Galliamov, Syed M Ahsan Kazmi, Adil Khan, Ad\'in Ram\'irez Rivera 2/17/2026

Concepts' Information Bottleneck Models

Information bottleneck regularizer for concept bottleneck models to improve interpretability while maintaining accuracy.

Ax Stefano Woerner, Seong Joon Oh, Christian F. Baumgartner 2/17/2026

Universal Algorithm-Implicit Learning

Theoretical framework for meta-learning defining practical universality and distinguishing algorithm-implicit learning capabilities.

Ax Julien Siems, Riccardo Grazzi, Kirill Kalinin, Hitesh Ballani, Babak Rahmani 2/17/2026

Learning State-Tracking from Code Using Linear RNNs

Converts state-tracking tasks from sequence-to-sequence to next-token prediction format for training language models with linear RNNs.

Ax Yu Huang, Zixin Wen, Yuejie Chi, Yuting Wei, Aarti Singh, Yingbin Liang, Yuxin Chen 2/17/2026

On the Learning Dynamics of RLVR at the Edge of Competence

Theoretical analysis of training dynamics in reinforcement learning with verifiable rewards for transformers on compositional reasoning tasks.

Ax Jivat Neet Kaur, Isaac Gibbs, Michael I. Jordan 2/17/2026

Locally Adaptive Multi-Objective Learning

Online learning approach for multi-objective prediction with adaptive algorithms handling arbitrary distribution shift and multiple simultaneous objectives.

Ax Xander Davies, Giorgi Giglemiani, Edmund Lau, Eric Winsor, Geoffrey Irving, Yarin Gal 2/17/2026

Boundary Point Jailbreaking of Black-Box LLMs

New jailbreak attack method (BPJ) that automatically evades classifier-based safeguards in frontier LLMs without requiring white/grey-box access.

Ax Subham Sekhar Sahoo, Jean-Marie Lemercier, Zhihan Yang, Justin Deschenaux, Jingyu Liu, John Thickstun, Ante Jukic 2/17/2026

Scaling Beyond Masked Diffusion Language Models

First scaling law study comparing masked diffusion and uniform-state discrete diffusion language models, showing masked diffusion performance characteristics.