Our team, VANGUARD, presents IROH (Insightful Ranking of Humor), a three-stage retrieval system for JOKER Task 1 English at CLEF 2026, achieving first place on the leaderboard with 0.6347 MAP. Our pipeline combines hybrid sparse-dense retrieval, cross-encoder reranking, and a LoRA-adapted Large Language Model judge ens...
A. Mocanu, Sebastian Mocanu, Ciprian-Octavian Truică et al.· 0 citations
The BabyLM challenge measures how much language a model can learn from developmentally-plausible, child-scale data rather than internet-scale corpora, yet prior language models forgo the biological constraints of the neural circuitry that acquires human language: spiking neurons separated into excitatory and inhibitory...
Multilingual Large Language Models (LLMs) traditionally rely on a single vocabulary shared by all supported languages, which can lead to uneven compression across them. Moreover, their large embedding and output matrices increase memory usage and slow inference, notably for small-scale models. It is also wasteful as mo...
Franck Signe, Hippolyte Pilchen, François Yvon et al.· 0 citations
The North Small Translate model is presented, an open-weight, LLM-based machine translation (MT) model with instruction-following capabilities built on the same foundation as Cohere's Command A Plus, a mixture-of-experts architecture with 25 billion active parameters out of 218 billion total parameters.
Tom Kocmi, Alexandre Berard, P. Blunsom et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Findings motivate variety-aware language identifiers, references, and reporting for cross-border languages or language dialects/variants for cross-border languages or language dialects/variants.
Felermino D. M. A. Ali, D. Mateus, M. Mangue· 0 citations
Large language models (LLMs) are increasingly used for emotional support and relationship advice, where a model's tendency to preserve a user's face can inadvertently reinforce harmful interpersonal behaviors. To systematically examine this risk, we developed the Romantic Relationship Advice-Seeking Prompts (RRASP) dat...
Helen Choi, Edric Castel C. Hao, K. Bautista et al.· 0 citations
This work focuses on machine translation models for Pacific creoles: Tok Pisin, Bislama, and Solomon Pijin, with particular attention to broad domain performance, and relies on a number of data collection and preparation techniques, including LLM-assisted respelling and alignment, back-translation, and distillation fro...
Raphael Merx, Nick Thieberger, Ekaterina Vylomova· 0 citations
It is shown that lightweight reference set distributional diagnostics can reveal model-specific instability patterns during sequential LoRA personalization of SLMs, including cases where task-level metrics alone hide harmful adaptation.
Thomas S. Paula, L. Kupssinskü, Rodrigo C. Barros· arXiv.org· 0 citations
Fixed-SAE Track is introduced, a framework that trains one shared SAE per considered layer on activations pooled across the base model and all RL checkpoints, holding every feature direction fixed so that representation shifts are rigorously defined through the activations of interpretable SAE latents, including the de...
Ling-Heng Du, Yi-Ming Tang, Xufeng Duan et al.· 1 citation
T-SMART is presented, a neurosymbolic framework that separates these roles: a frozen LLM interprets questions and selects operations, deterministic tools perform numerical computation, and structured perception is invoked only when needed to provide a controlled framework for analyzing component contributions in neuros...
I. Delgado, Himansi Gupta, B. Khatri et al.· 0 citations
Predictive likelihood ratios that average over uncertain probability deficits and residual-tail distributions are constructed, showing robustness across the evaluated alternatives, not uniform power guarantees or resistance to arbitrary text edits.
Mechanistic interpretability has become the dominant lens for understanding frontier language models, as their inner workings are complex and inherently black boxes. To gain insights into these models, interpretable replacement networks (IRNs) are trained at all layers, exposing interpretable features through sparsely...
StudentSim is a research system for evaluating and improving AI tutoring tools before they are tested with learners. It models how individuals may engage with tutors, giving feedback to improve tutor performance across subjects like chess, writing, and mathematics. The post Simulating student learning for better AI tutoring appeared first on Microsoft Research.
