A fundamental gap between generation fluency and reasoning ability is uncovered, a pronounced ''inverse scaling effect'' where larger models can underperform in domain-specific reasoning, and systemic bottlenecks across all SOTA models are identified, exposing fundamental limitations of current architectures.
Guangtao Nie, Huimu Wang, Gewei Lu et al.· Proceedings of the 32nd ACM...· 0 citations
This work proposes Multi-Marginal Preference Optimization (MMPO), a fine-grained framework that intervenes at the data, gradient, and constraint levels rather than relying on coarse-grained global scalarization to address optimization conflicts among multiple objectives in real-world deployment scenarios.
Shang-Wen Mao, Hao Zhang, Guangtao Nie et al.· 1 citation· ⚡1
A fundamental gap between generation fluency and reasoning ability is uncovered, a pronounced ''inverse scaling effect'' where larger models can underperform in domain-specific reasoning, and systemic bottlenecks across all SOTA models are identified, exposing fundamental limitations of current architectures.
Guangtao Nie, Huimu Wang, Gewei Lu et al.· Proceedings of the 32nd ACM...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
