Almanac
model

LLaDA

modelactiveprovisionalllada-4b486dd3·2 events·first seen 16d ago

Aliases: LLaDA

Co-occurring entities

More like this (12)

Recent events (2)

6arXiv · cs.CL·16d ago·source ↗

Trajectory Analysis of Masked Diffusion LMs for Graph-to-Text Generation with Lambda-Scaled Structural Decoding

This paper presents the first systematic study of masked diffusion language models (MDLMs) for graph-to-text generation, analyzing the order in which tokens are unmasked during iterative decoding. The authors find MDLMs naturally unmask entities first, then relational/function words, then structural tokens—a pattern disrupted by supervised fine-tuning, which prematurely anchors structural tokens and causes hallucination or omission. They propose lambda-scaled structural decoding, a training-free inference-time fix that recovers +9.4 BLEU-4, and introduce Graph-LLaDA, which integrates a Graph Transformer encoder into LLaDA's decoding process. Cross-dataset evaluation on the LAGRANGE benchmark shows prior baselines overfit to dataset-specific patterns while MDLM-based approaches generalize better.

5arXiv · cs.CL·13d ago·source ↗

Knowledge editing via locate-then-edit transferred to masked diffusion language models, revealing multi-token failure mode

A new arXiv paper investigates whether locate-then-edit knowledge editing methods, developed for autoregressive models, transfer to masked diffusion language models (MDMs) such as LLaDA and Dream. The authors find that causal tracing identifies the same early-to-mid-layer MLP location in both paradigms, but MDMs degrade systematically on multi-token edits due to partially unmasked intermediate states that the edit was never optimized for. A correction targeting these intermediate states substantially restores multi-token editing performance. The work is the first systematic comparison of knowledge editing across autoregressive and diffusion-based language model paradigms.