MDM-Prime-v2 is a masked diffusion language model that improves upon the original MDM-Prime by addressing two key limitations: the high cross‑entropy loss caused by common Byte‑Pair‑Encoding tokenizers and the lack of guidance for choosing token granularity. The new design incorporates Binary Encoding and Index Shuffling, and the authors provide an analysis linking token granularity and sub‑token entropy to training objectives and downstream performance. When scaled to 1.1 B parameters, MDM-Prime‑v2 achieves higher average zero‑shot accuracy on eight commonsense reasoning benchmarks than comparable models such as GPT‑Neo, OPT, Pythia, Bloom, SMDM, and TinyLLaMA.
arXiv:2605. 13026v2 Announce Type: replace-cross Abstract: Masked diffusion models (MDMs) have emerged as a promising alternative to autoregressive models (ARMs) for language modeling.
arXiv:2609. 04010v1 Announce Type: new Abstract: Large Language Models (LLMs) owe much of their success to next-token prediction (NTP), but their autoregressive (AR) structure requires slow, sequential token generation.
arXiv:2607. 15200v1 Announce Type: cross Abstract: Reinforcement learning has proven effective for improving reasoning in large language models, but extending it to Masked Diffusion Language Models (MDLMs) remains challenging due to the intractability of the log-likelihood estimation.
arXiv:2609.38385v1 Announce Type: new Abstract: Supervised fine-tuning of discrete diffusion language models masks some response tokens and trains the model to recover their original values from the...
arXiv:2601. 22954v2 Announce Type: replace-cross Abstract: Diffusion Large Language Models (dLLMs) have emerged as a promising alternative to purely autoregressive language models because they can decode multiple tokens in parallel.
The paper introduces Less Uniform Diffusion (LUDI), a framework that improves uniform diffusion language models by using a less uniform loss and per-token time embeddings to guide reverse transitions and enable confidence-based few-step sampling. Experiments demonstrate that LUDI provides cleaner supervision, enhances few-step generation, and scales to a 7B model (LUDI-7B) that achieves a 3-token-per-step speedup over autoregressive decoding while matching masked diffusion baselines. The work suggests that UDLMs still have untapped potential for complex generation tasks.
The paper investigates whether small models distilled from larger ones behave similarly when using byte versus token tokenization. It introduces two methods—Marginalize‑It (approximate) and End‑Of‑Token (exact)—to convert token logits to byte logits, and conducts a large‑scale study on decoder‑only dense transformers ranging from 1 billion to 1 trillion bytes of data. Results show that while token‑based models excel early, byte‑based models eventually surpass them with more compute, achieving higher performance ceilings, greater data efficiency, and lower logit storage costs.
arXiv:2505. 18227v4 Announce Type: replace-cross Abstract: In Transformer architectures, tokens\textemdash discrete units derived from raw data\textemdash are formed by segmenting inputs into fixed-length chunks.
The paper introduces TokenAdapt, a model‑agnostic tokenizer transplantation method that uses a hybrid heuristic to initialize new token embeddings, and a novel pre‑tokenization learning approach for multi‑word Supertokens to improve compression. TokenAdapt combines local subword decomposition and global semantic similarity to preserve semantics while reducing retraining needs. Empirical results show that TokenAdapt outperforms existing baselines such as Transtokenizer and ReTok, achieving lower perplexity ratios and significant compression gains.
The paper introduces two new tokenisation algorithms—BottomUpLL and TopDownComp—to systematically explore the 2x2 design space defined by optimisation objective (compression vs. log‑likelihood) and search procedure (bottom‑up merging vs. top‑down pruning). Experiments across model sizes, vocabularies, and domains show that the search procedure, rather than the objective, consistently yields lower bits‑per‑byte, while no clear pattern emerges on the BLiMP benchmark. These findings clarify how tokeniser design choices influence language‑model performance and provide guidance for constructing tokenisers more principledly.
arXiv:2608. 03457v1 Announce Type: new Abstract: Diffusion language models (dLLMs) offer an alternative to autoregressive (AR) language modeling, yet the scaling behavior of Mixture-of-Experts (MoE) dLLMs remains poorly understood.
arXiv:2601. 22450v2 Announce Type: replace-cross Abstract: Masked Diffusion Language Models have recently emerged as a powerful generative paradigm, yet their generalization properties remain understudied compared to their auto-regressive counterparts.