Neural & Evolutionary

Ivan Sviridov, Artem Oskin, Ivan Panin et al.

For medical AI practitioners, this work provides a zero-shot, fine-tuning-free approach to optimize LLM-based clinical pipelines, though the gains are demonstrated on specific benchmarks.

Guy Kaplan, Zorik Gekhman, Zhen Zhu et al.

For practitioners fine-tuning LLMs, this work provides a practical method to reduce hallucinations while learning new facts, addressing a critical reliability issue.

Susung Hong, Brian Curless, Ira Kemelmacher-Shlizerman et al. · uw

This addresses the challenge of automated creative content production for entertainment, though it appears incremental as it builds on existing agent and LLM methods.

Ahmadreza Jeddi, Minh Ngoc Le, Hakki C. Karaimer et al.

For developers of autonomous research agents, GEAR demonstrates that maintaining multiple promising directions and adapting search strategy over time improves effectiveness.

Haoze Lv, Ning Lu, Ziang Zhou et al.

This work addresses the inefficiency of existing LLM-based automatic heuristic design frameworks by enabling proactive decision-making, which is significant for researchers and practitioners tackling NP-hard combinatorial optimization problems.

Nico Pelleriti, Sree Harsha Nelaturu, Zhanke Zhou et al.

For researchers developing and evaluating evolutionary coding agents, this work provides a diagnostic methodology to distinguish between different mechanisms behind performance gains, revealing that final benchmark scores can be misleading.

Xinhao Zhang, Xi Chen, François Portet et al.

Provides actionable insights for designing and training LLM-based optimization systems by analyzing optimization trajectories across 15 LLMs and 8 tasks.

Michal Olak, Tommaso Boccato, Matteo Ferrante

This work addresses speech brain-computer interfaces for individuals with speech impairments, offering incremental improvements in decoding accuracy and robustness.

Ryan Bahlous-Boldi, Isha Puri, Idan Shenfeld et al.

For LLM post-training, VPO addresses the bottleneck of low-entropy responses that hinder test-time search, offering a drop-in replacement for GRPO that improves search performance.

Albert Zeyer, Tim Posielek, Ralf Schlüter et al.

This work provides practical guidance for researchers and practitioners building faster speech recognition systems by simplifying text-data integration.

Rui Tang, Kaiyu Xu, Pengsen Cheng et al.

For LLM safety researchers, EvoJail provides a more adaptive and diverse automated jailbreak generation method to uncover safety weaknesses across evolving models.

Davyd Naveriani, Albert Zeyer, Ralf Schlüter et al.

For speech recognition researchers, this provides a new approach to rescoring and decoding that leverages diffusion models for better accuracy.

David McAllister, Miika Aittala, Tero Karras et al.

This work addresses a specific bottleneck in RL-based fine-tuning for diffusion models, offering incremental improvements for researchers and practitioners in image synthesis.

Lakshya A Agrawal, Donghyun Lee, Shangyin Tan et al.

Demonstrates that text optimization with LLM-based search is a general-purpose problem-solving paradigm, unifying tasks traditionally requiring domain-specific algorithms.

Joel Z. Leibo, Alexander Sasha Vezhnevets, Manfred Diaz et al.

This offers a novel theoretical framework for cognitive science and social norms, potentially redefining rationality, but it is incremental as it builds on existing models like LLMs.

Ethan Caballero, Priyank Jaini, David Krueger et al.

Provides a more accurate tool for predicting performance of large-scale neural networks across diverse domains, aiding resource allocation and architecture design.

Jacob Fein-Ashley, Paria Rashidinejad

For practitioners of large-scale language models and reasoning systems, Attractor Models offer a scalable way to incorporate iterative refinement without the overhead of deep recurrence.

Shaofeng Zhang, Shengcai Liu, Ning Lu et al.

This work addresses the problem of inefficient heuristic design in combinatorial optimization for researchers and practitioners by introducing an instance-aware approach, though it is incremental as it builds on existing LLM and evolutionary algorithm integrations.

Sam Earle, Kay Arulkumaran, Andrew Dai et al.

For AI researchers studying open-endedness, this work provides an initial replication and analysis of a canonical open-ended system with VLMs, but the results are preliminary and incremental.

Sohan Shankar, Yi Pan, Hanqi Jiang et al.

It proposes an integrative agenda for researchers in neuroscience, AI, and hardware to address critical challenges in developing brain-inspired AGI systems, though it is incremental as a survey and position paper.