Showing cs.CLShow all
2 papers · 1 filter
cs.CL2026
GAW-PO: Preference Optimization with Gradient-Aligned Token Weights
Andreea Dutulescu, Stefan Ruseti, Mihai Masala +2
Most preference optimization methods, such as Direct Preference Optimization (DPO), apply preference supervision at the response level, although autoregressive language models are…
cs.CL2026
Value-Aware Numerical Representations for Transformer Language Models
Andreea Dutulescu, Stefan Ruseti, Mihai Dascalu
Transformer-based language models often achieve strong results on mathematical reasoning benchmarks while remaining fragile on basic numerical understanding and arithmetic operatio…