[LG] Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach J Geiping, S McLeish, N Jain, J Kirchenbauer... [Max-Planck Institute for Intelligent System & University of Maryland] (2025)

[LG] On the Difficulty of Constructing a Robust and Publicly-Detectable Watermark J Fairoze, G Ortiz-Jiménez, M Vecerik, S Jha... [UC Berkeley & Google DeepMind] (2025)

[CL] Sparse Autoencoders for Hypothesis Generation R Movva, K Peng, N Garg, J Kleinberg... [UC Berkeley & Cornell University] (2025)

[LG] Training Language Models to Reason Efficiently D Arora, A Zanette [CMU] (2025)

[CL] DuoGuard: A Two-Player RL-Driven Framework for Multilingual LLM Guardrails Y Deng, Y Yang, J Zhang, W Wang... [University of California, Los Angeles] (2025)

[CL] When One LLM Drools, Multi-LLM Collaboration Rules S Feng, W Ding, A Liu, Z Wang... [University of Washington & The University of Texas at Austin & Google] (2025)

[CL] Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoning D Su, H Zhu, Y Xu, J Jiao... [Meta AI & UC Berkeley] (2025)

[CL] Achieving Operational Universality through a Turing Complete Chemputer D Gahler, D Thomas, S Lach, L Cronin [University Avenue] (2025)

[LG] Language Models Use Trigonometry to Do Addition S Kantamneni, M Tegmark [MIT] (2025)

[LG] Loss Functions and Operators Generated by f-Divergences V Roulet, T Liu, N Vieillard, M E. Sander... [Google DeepMind] (2025)

[LG] Fast Solvers for Discrete Diffusion Models: Theory and Applications of High-Order Algorithms Y Ren, H Chen, Y Zhu, W Guo... [Stanford University] (2025)

[LG] AI-driven materials design: a mini-review

[LG] Universal Sparse Autoencoders: Interpretable Cross-Model Concept Alignment H Thasarathan, J Forsyth, T Fel, M Kowal... [EECS York University,] (2025)

[LG] Harmonic Loss Trains Interpretable AI Models D D. Baek, Z Liu, R Tyagi, M Tegmark [MIT] (2025)

[LG] Gold-medalist Performance in Solving Olympiad Geometry with AlphaGeometry2 Y Chervonyi, T H. Trinh, M Olšák, X Yang... [Google DeepMind] (2025)

[LG] Sample, Scrutinize and Scale: Effective Inference-Time Search by Scaling Verification E Zhao, P Awasthi, S Gollapudi [Google Research] (2025)

[LG] Do Large Language Model Benchmarks Test Reliability? J Vendrow, E Vendrow, S Beery, A Madry [MIT] (2025)

[LG] Decision Trees That Remember: Gradient-Based Learning of Recurrent Decision Trees with Memory S Marton, M Schneider [University of Mannheim & Boehringer Ingelheim] (2025)

[LG] Great Models Think Alike and this Undermines AI Oversight S Goel, J Struber, I A Auzina, K K Chandra... [ELLIS Institute Tubingen & Tubingen AI Center] (2025)

[CL] LLM Alignment as Retriever Optimization: An Information Retrieval Perspective B Jin, J Yoon, Z Qin, Z Wang... [Google Cloud] (2025)