Tag
DiffIE introduces a diffusion-based method for open information extraction that uses stochastic reverse-diffusion to generate multiple candidate triplets, achieving state-of-the-art performance on benchmarks like CaRB and BenchIE.
STAR-OPD is a novel on-policy distillation method for ABSA quadruple extraction that uses set-structured rewards to correct structural errors in student models, improving performance and narrowing the gap with teacher models.
StepReflect reformulates per-step GUI reflection for mobile agents as supervised structured prediction, achieving higher transition accuracy than GPT-5.2 on AndroidWorld while reducing API costs.
This paper introduces Approximate Structured Diffusion, a method that combines conditional random fields (CRFs) with discrete diffusion for sequence labelling. It uses a CRF conditioned on noisy label sequences and approximate mean-field inference, achieving a 16.5% error reduction on POS tagging.
The article introduces Needle, a 26M parameter model by Cactus-Compute designed for single-shot tool calling, arguing that tool routing should be separated from reasoning as a structured prediction task to improve agent efficiency and latency.