SRCB at SemEval-2026 Task 5: A Multi-Target Finetuning Framework for Large Language Models with Joint Regression and Text Generation

The Interlingua Hypothesis: LLMs Translate via a Latent Task-agnostic Feature Space

Three lines of evidence are shown that provide convergent evidence in support of the interlingua hypothesis, which holds that language models translate by reading a source sentence into a latent feature space, and generate a target sentence by reading from the latent feature space.

Unveiling the semantic potential of BERT for unsupervised Chinese WSD via non-linear semantic sharpening

This paper proposes a training-free inference-stage enhancement termed Semantic-Sharpened Masked Language Modeling, which achieves performance comparable to knowledge-enhanced models without additional training or structural modifications, thereby improving the balance between overall accuracy and sense-level discrimin...

Team SINAI at PAN 2026: Combining Stylometric Embeddings and IDF Features for Generative AI Text Detection

Think Before You Transcribe: Context-Steered Reasoning for Hotword-Aware Speech Recognition

Contextual automatic speech recognition (ASR) improves rare-entity recognition by supplying candidate hotwords, but retrieved candidates may be incomplete or incorrect. A reliable system must therefore exploit a relevant candidate without allowing unsupported alternatives to override the acoustic evidence. We propose a...

ArguMind at Touché: A Unified Multi-Task Transformer Approach for Fallacy and Argument Scheme Classification

TTLab at StanceEval-2026: A Cloze-Style Prompting Approach for Arabic-Language Stance Detection (CLASP-Ar)

This work introduces $\texttt{CLASP-Ar}$, which reformulates the Arabic-language stance detection task as cloze-style masked language modeling, and reformulates the task as cloze-style masked language modeling.