Researchers Develop Continuous-Time Markov Chain Framework for Insertion Language Models
Researchers have proposed a continuous-time Markov chain (CTMC) framework that derives a diffusion-style denoising objective for Insertion Language Models (ILMs) from first principles. Prior ILM formulations were largely ad-hoc, and this work unifies them as special cases of a single denoising framework. The approach offers competitive language modeling performance alongside greater sampling flexibility, addressing a theoretical gap in non-left-to-right text generation.
A paper accepted at AISTATS 2026 introduces a principled framework for Insertion Language Models by modeling the noising process as a continuous-time Markov chain over variable-length sequences. This formulation yields a diffusion-style denoising objective derived from first principles, in contrast to the ad-hoc designs that have characterized previous ILM work. The authors demonstrate that existing ILM approaches can be recovered as special cases within this unified framework. Empirical evaluation on a synthetic planning task shows the method retains the known advantages of insertion-based generation over both left-to-right autoregressive models and masked diffusion models. On standard language modeling benchmarks, the diffusion-based ILM is competitive with these baselines while offering additional flexibility in how sequences are sampled. Code has been made publicly available alongside the paper.
What's missing
The paper evaluates language modeling competitiveness against baselines but does not report results on large-scale, real-world NLP benchmarks; it is unclear how the approach scales to modern large language model regimes. The synthetic planning task, while illustrative, may not fully capture the complexity of practical generation scenarios. Additionally, computational cost and inference speed relative to autoregressive and masked diffusion baselines are not discussed in the abstract.
What different sources said
- arXiv cs.LGCenter
A Continuous-Time Markov Chain Framework for Insertion Language Models
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.