EPO: Explicit Policy Optimization for Strategic Reasoning in LLMs via Reinforcement Learning

Posted by dbos Peking

Kavli Affiliate: Ke Wang | First 5 Authors: Xiaoqian Liu, Ke Wang, Yongbin Li, Yuchuan Wu, Wentao Ma | Summary: Large Language Models (LLMs) have shown impressive reasoning capabilities in well-defined problems with clear solutions, such as mathematics and coding. However, they still struggle with complex real-world scenarios like business negotiations, which require strategic reasoning-an […]


Continue.. EPO: Explicit Policy Optimization for Strategic Reasoning in LLMs via Reinforcement Learning

EPO: Explicit Policy Optimization for Strategic Reasoning in LLMs via Reinforcement Learning

Posted by dbos Peking

Kavli Affiliate: Ke Wang | First 5 Authors: Xiaoqian Liu, Ke Wang, Yongbin Li, Yuchuan Wu, Wentao Ma | Summary: Large Language Models (LLMs) have shown impressive reasoning capabilities in well-defined problems with clear solutions, such as mathematics and coding. However, they still struggle with complex real-world scenarios like business negotiations, which require strategic reasoning-an […]


Continue.. EPO: Explicit Policy Optimization for Strategic Reasoning in LLMs via Reinforcement Learning

EquiBench: Benchmarking Code Reasoning Capabilities of Large Language Models via Equivalence Checking

Posted by dbos Peking

Kavli Affiliate: Ke Wang | First 5 Authors: Anjiang Wei, Jiannan Cao, Ran Li, Hongyu Chen, Yuhui Zhang | Summary: Equivalence checking, i.e., determining whether two programs produce identical outputs for all possible inputs, underpins a broad range of applications, including software refactoring, testing, and optimization. We present the task of equivalence checking as a […]


Continue.. EquiBench: Benchmarking Code Reasoning Capabilities of Large Language Models via Equivalence Checking

EquiBench: Benchmarking Code Reasoning Capabilities of Large Language Models via Equivalence Checking

Posted by dbos Peking

Kavli Affiliate: Ke Wang | First 5 Authors: Anjiang Wei, Jiannan Cao, Ran Li, Hongyu Chen, Yuhui Zhang | Summary: Equivalence checking, i.e., determining whether two programs produce identical outputs for all possible inputs, underpins a broad range of applications, including software refactoring, testing, and optimization. We present the task of equivalence checking as a […]


Continue.. EquiBench: Benchmarking Code Reasoning Capabilities of Large Language Models via Equivalence Checking

Dominant Role of Coplanar Inflows in Driving Disk Evolution Revealed by Gas-Phase Metallicity Gradients

Posted by dbos Peking

Kavli Affiliate: Yingjie Peng | First 5 Authors: Cheqiu Lyu, Enci Wang, Hongxin Zhang, Yingjie Peng, Xin Wang | Summary: Using spatially resolved spectroscopic data from the MaNGA sample, we investigate the parameters influencing the radial gradients of gas-phase metallicity ($nablalog(mathrm{O/H})$), to determine whether disk formation is primarily driven by coplanar gas inflow or by […]


Continue.. Dominant Role of Coplanar Inflows in Driving Disk Evolution Revealed by Gas-Phase Metallicity Gradients

LanP: Rethinking the Impact of Language Priors in Large Vision-Language Models

Posted by dbos Berkeley

Kavli Affiliate: Xiang Zhang | First 5 Authors: Zongyu Wu, Yuwei Niu, Hongcheng Gao, Minhua Lin, Zhiwei Zhang | Summary: Large Vision-Language Models (LVLMs) have shown impressive performance in various tasks. However, LVLMs suffer from hallucination, which hinders their adoption in the real world. Existing studies emphasized that the strong language priors of LVLMs can […]


Continue.. LanP: Rethinking the Impact of Language Priors in Large Vision-Language Models

Single-cell multiome and spatial profiling reveals pancreas cell type-specific gene regulatory programs driving type 1 diabetes progression

Posted by dbos Hopkins

Kavli Affiliate: Michael Miller | Authors: Rebecca Melton, Sara Jimenez, Weston Elison, Luca Tucciarone, Abigail Howell, Gaowei Wang, Denise Berti, Elisha Beebe, Michael Miller, Chun Zeng, Kennedy Vanderstel, Katha Korgaonkar, Ruth Elgamal, Hannah Mummey, Josh Chiou, Emily Griffin, Irina Kusmartseva, Mark A. Atkinson, Sebastian Preissl, Fabian Theis, Maike Sander and Kyle J Gaulton | Summary: […]


Continue.. Single-cell multiome and spatial profiling reveals pancreas cell type-specific gene regulatory programs driving type 1 diabetes progression

Learning decouples accuracy and reaction time for rapid decisions in a transitive inference task

Posted by dbos Columbia

Kavli Affiliate: Vincent Ferrera | Authors: Fabian A Munoz Silva, Greg Jensen, Maxwell Shinn, Yelda Alkan, John Murray, Herbert Terrace and Vincent P Ferrera | Summary: The accumulation of evidence over time formalized in the drift diffusion model (DDM), has become one of the most prevalent models of deliberative decision-making. To better understand the role […]


Continue.. Learning decouples accuracy and reaction time for rapid decisions in a transitive inference task

The life cycle of giant molecular clouds in simulated Milky Way-mass galaxies

Posted by dbos MIT

Kavli Affiliate: Mark Vogelsberger | First 5 Authors: Yang Ni, Hui Li, Mark Vogelsberger, Laura V. Sales, Federico Marinacci | Summary: In this work, we trace the complete life cycle of individual GMCs in high-resolution Milky Way-mass galaxy simulations to determine how different stellar feedback mechanisms and galactic-scale processes govern cloud lifetimes, mass evolution, and […]


Continue.. The life cycle of giant molecular clouds in simulated Milky Way-mass galaxies

JADES: Average Nitrogen Enhancement in High-Redshift Broad-Line Active Galactic Nuclei

Posted by dbos Cambridge

Kavli Affiliate: Roberto Maiolino | First 5 Authors: Yuki Isobe, Roberto Maiolino, Francesco D’Eugenio, Mirko Curti, Xihan Ji | Summary: The unexpectedly high nitrogen-to-oxygen (N/O) ratios observed in high-redshift (z) galaxies have challenged our understanding of early star formation. Notably, many of these nitrogen-rich galaxies show signatures of active galactic nuclei (AGNs), suggesting a possible […]


Continue.. JADES: Average Nitrogen Enhancement in High-Redshift Broad-Line Active Galactic Nuclei