PRISM is a framework that evaluates how activation interventions transfer across precision, languages, and model scales to mitigate unsafe tool‑calling behaviors in language models. By leveraging lower‑resource settings such as quantized NF4 labels and cross‑lingual transfer, PRISM demonstrates that alignment improvements can be achieved with minimal scenarios and that these gains transfer to held‑out languages and larger models. The study shows that auxiliary supervision from compressed models and multilingual data can effectively uncover and reduce hazardous tool‑calling behaviors across different model families.
By Giovana Kerche Bon\'as, Thales Sales Almeida, Marcelo da Silva Reis, Rodrigo Nogueira, Marcos Meideiros Raimundo
arXiv:2608. 09521v1 Announce Type: new Abstract: Activation-based tools are usually tied to one model's native hidden space, requiring probes, sparse autoencoders, and natural-language interpreters to be rebuilt or rediscovered for each new language model.
By Su-Hyeon Kim, Jiwan Mun, Yo-Sub Han
arXiv:2610.19696v1 Announce Type: new
Abstract: Multi-agent systems allow language models to solve tasks collaboratively by sharing complementary information. These systems typically communicate thro...
By Xin Li, Mengbing Liu, Hao Jiang, Xin Gao, Annan Wang, Xingwei Qu, Yichi Zhang, Chau Yuen
The paper proposes a lightweight recurrent memory module inserted between the lower and upper halves of a 6‑layer decoder‑only transformer. This module, which uses cross‑attention to observe hidden states, a GRU to update a persistent state, and gated addition to modulate subsequent layers, adds only 3.7% more parameters. It reduces evaluation loss by 28.5% and narrows the generalization gap, with ablations showing the benefit comes solely from the memory topology rather than auxiliary losses.
By Eduardo Novaes Hering
Domain fine-tuning degrades the safety of large language models: fine-tuned specialists readily comply with harmful prompts framed in domain language. Existing inference-time defenses that mix logits from a safe anchor model require both models to share a vocabulary, which rules them out for the cross-family specialists where safety is most degraded.
The paper introduces a method for determining whether retrieval-augmented generation (RAG) systems have sufficient, insufficient, or conflicting evidence to answer a question. By training a lightweight linear classifier on hidden activations and attention-derived features from 16 language models, the authors demonstrate that these internal signals reliably predict the adequacy of retrieved documents, outperforming prompting-based baselines and specialized RAG models. Analysis shows that middle-layer hidden states carry the most informative signals for this triage task.
By Syed Mahbubul Huq, Christopher Child, Tillman Weyde, Pranava Madhyastha
arXiv:2606. 12342v1 Announce Type: cross Abstract: Domain fine-tuning degrades the safety of large language models: fine-tuned specialists readily comply with harmful prompts framed in domain language.
By Chirag Chawla, Pratinav Seth, Vinay Kumar Sankarapu
arXiv:2602. 12952v3 Announce Type: replace Abstract: Adapting large pre-trained models to downstream tasks often produces task-specific parameter updates that are expensive to relearn for every model variant.
By Filippo Rinaldi, Aniello Panariello, Giacomo Salici, Angelo Porrello, Simone Calderara
arXiv:2607. 11228v1 Announce Type: cross Abstract: While Large Vision-Language Models (LVLMs) demonstrate remarkable capabilities, they remain highly susceptible to embedded social biases.
By Anqi Li, Jie Zhang, Zhongqi Wang, Songkai Xue, Jiahao Wang, Shiguang Shan, Xilin Chen
arXiv:2605.30018v3 Announce Type: replace-cross
Abstract: Large language models (LLMs) frequently achieve impressive scores on standardized benchmarks, yet accuracy alone offers a limited view of the...
By Tanmoy Chakraborty, Ayan Sengupta, Suparna Bhattacharya, Partha Pratim Chakrabarti, Amlan Chakrabarti, Supratik Chakraborty, Partha Pratim Das, Lipika Dey, Richa Singh, Mayank Vatsa
The paper investigates how the choice of Transformer layers for fine‑tuning affects both task performance and the preservation of pre‑trained capabilities. By measuring layer sensitivity with a forward‑only input–output cosine similarity metric, the authors identify layers that are more suitable for adaptation. They introduce Layer‑Selective LoRA (LS‑LoRA), which trains adapters only in low‑similarity layers, and demonstrate that this approach improves target‑task results while better retaining commonsense reasoning compared to standard all‑layer LoRA.
By Zhiqiang Pang, Zihong Sun, Qi Xie, Jun Shu, Deyu Meng, Zongben Xu
arXiv:2610.34756v1 Announce Type: cross
Abstract: Low-Rank Adaptation (LoRA) has emerged as the dominant paradigm for parameter-efficient fine-tuning of large language models. Standard all-layer LoRA...
By Fei Zuo, Yizhou Huang, Kezhi Wang