arXiv Machine Learning
Relative Representations Enable Portable White-Box Monitoring Across Language Models
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Computation and Language
Oct 16
PRISM: Transfer and Limits of Activation Interventions in Tool-Calling Language Models
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.
arXiv AI
Aug 11
One Adapter Pair per Model: A Universal Activation Interface for Language Models
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.
arXiv Computation and Language
Oct 15
Share the Symbol, Not the Vector: Training-Free Fact Communication Across Language Models
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...
arXiv Computation and Language
Sep 16
Persistent Recurrent Memory Between Transformer Layers - Improves Language Model Generalization
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.
Hugging Face Trending Papers
Jun 10
ALIGNBEAM : Inference-Time Alignment Transfer via Cross-Vocabulary Logit Mixing
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.
arXiv Computation and Language
Aug 31
Knowing Before Answering: Decoding Language Models for Reliable RAG
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.