ArtificialIntelligence.io

The Signal

Everything that matters in AI, with our take.

Updated through the day. Every headline links straight to the source. The two lines underneath are ours.

arXiv cs.CLPaper

Data Scarcity and Model Sparsity: Mixtures-of-Experts Overfit More to Repeated Data

As text becomes scarce, data repetition is standard practice. This paper shows MoE architectures suffer disproportionately, losing their efficiency advantage around 4x repetition where dense models hold steady until 8x. If you're training sparse models at scale on limited unique data, this suggests dense models might compete better than conventional wisdom says. The hidden message: sparsity has a cost when data is constrained.

arXiv cs.AIPaper

ACE: Adaptive Calibration-Free Expert Skipping for MoE-based LLMs

MoE routing is stuck routing the same number of experts per token, which wastes compute on tokens that don't need it. This paper's contribution is narrow: a calibration-free algorithm to decide which experts to skip. Real but incremental. It matters if you're running MoE inference at scale, but the gains are meaningful only if expert-skipping wasn't already on your roadmap.

arXiv cs.CLPaper

Can Large Language Models Anticipate Behavioral Responses to Social Policies? A Case of Pension Enrollment Prediction among China's Flexible Workers

This demonstrates LLMs can function as policy simulation tools when domain-specialized and fine-tuned with causal context. The technique—anchoring prompts with econometric signals then distilling into a smaller model—is reusable for other policy-domain applications. Worth studying if you're building systems that need to predict behavioral responses to rule changes.

arXiv cs.CLPaper

PCoMoE: Shifting MoE Inference from Monolithic Expert Selection to Fine-Grained Path Composition

MoE models are becoming standard for efficient LLM inference, and the whole-expert abstraction is an optimization boundary that's too coarse. PCoMoE decomposes experts into reusable sub-expert structures and prunes low-value paths, which means cheaper inference. If you're serving MoE models (like Meta's models at scale) or considering whether to adopt MoE, this technique directly reduces your compute cost. The 1.31x speedup on hardware-constrained inference is real leverage.

arXiv cs.CLPaper

Beyond Magnitude: Contrastive Routing for Modular Mixture-of-Experts

A clean architectural insight with real empirical gains: contrasting against shared state instead of absolute magnitude makes expert boundaries align with linguistic structure and delivers 0.67-1.77 point accuracy improvements across nine benchmarks. This is relevant if you're shipping MoE models or evaluating them. The gains are modest but consistent and the mechanism is sound. If you control your router design, this is worth testing.

arXiv cs.LGPaper

DeaMoE: Efficient MoE Structure for Fast Small-Batch Decoding

This matters if you're serving MoE models in latency-critical applications. Small-batch inference is the real constraint in production agent systems and coding assistants, and expert-weight loading is a real bottleneck this paper addresses with a clean architectural idea: parameter sharing within expert groups. The tradeoff is accuracy versus speed; that data isn't clear in the excerpt. Check the numbers before adopting.