Beyond External Monitors: Enhancing Transparency of Large Language Models for Easier Monitoring

Guanxu Chen, Jing Shao, Tao Luo, Lijie Hu, Qihao Lin, Dongrui Liu

TL;DR

Instead of relying on passive external tools to monitor complex, entangled LLM representations, this paper introduces TELLME (Transparency Enhancement of LLMs without External modules). By actively reorganizing the internal representation space via contrastive learning, TELLME makes LLMs inherently easier to monitor and safer, achieving significant gains in both monitorability and self-detoxification across various model scales and modalities.

What Problem Does This Paper Address?

As Large Language Models (LLMs) grow more powerful, understanding their internal decision-making processes becomes critical for safety. Traditional safety frameworks rely on external monitors to screen inputs/outputs or use post-hoc interpretability tools like Sparse Autoencoders (SAEs) to decode internal states.

However, these approaches are inherently passive. If an LLM's internal features are deeply entangled, external tools struggle to decode them effectively, often incurring massive computational costs. Furthermore, behavioral methods like Chain-of-Thought (CoT) prompting can be unfaithful or even deceptive. This paper tackles a fundamental question: Can we actively modify the LLM itself to make its internal thinking process cleaner and easier to monitor?

Methods

The authors propose TELLME, a method that directly reshapes the LLM's latent representation space during post-training using Low-Rank Adaptation (LoRA). Instead of altering the architecture, TELLME fine-tunes a target layer (typically at 80% depth) using a joint objective function:

  1. Disentangle Loss ($\mathcal{L}_d$): Utilizing a classical InfoNCE contrastive loss, this component constructs positive pairs (similar behaviors, like two different examples of "bias") and negative pairs (distinct behaviors, like "violence" vs. "deception"). It maximizes representation similarity for similar concepts and minimizes it for distinct ones.
  2. Retain Loss ($\mathcal{L}_r$): To ensure the model doesn't lose its general capabilities or conversational stability, TELLME enforces an $\ell_2$ norm constraint on intermediate representations and a KL divergence penalty on output probabilities against the original unaltered model.
Figure 1. TELLME is designed to enhance the transparency of LLMs and makes them easier to monitor without external modules. Disentanglement of different behaviors in LLMs’ representation space improve their transparency, achieving better monitoring reliability and safety performance.

Datasets

The framework relies on two types of datasets:

  • Retain Sets (General Capabilities): UltraChat for text models and RLHF-V for multimodal models.
  • Disentangle & Evaluation Sets (Safety/Risks): BeaverTails (for Multi-Risk and Safe-or-Harmful tasks), SPA-VL (for multimodal safety), and a dataset from Taylor et al. (2025) for detecting reward-hacking misalignment.
  • General Benchmarks: MMLU, MATH, CoLA (GLUE), and SiQA were used to verify that general capabilities remained intact.

Key Insights and Ideas

  • Active vs. Passive Monitoring: Actively structuring the internal representation space is far more efficient than training massive external feature extractors on top of a messy, unaltered latent space.
  • The "Hidden Bonus" of Self-Detoxification: Simply separating safe and harmful behaviors in the representation space automatically improves the model's safety generation. Crucially, this happens without explicit preference data (like DPO) or explicitly instructing the model on which behavior is preferred.
  • Theoretical Alignment with Optimal Transport: The authors ground TELLME's success in optimal transport theory. They demonstrate that aggregating similar behaviors reduces the k-variance of the behavior distributions, which mathematically minimizes the generalization bound for behavior classification.
Figure 2. Overview of TELLME. TELLME disentangles representations by maximizing the examples’ similarities of similar behaviors, and minimizing the examples’ similarities of different behaviors. Meanwhile, TELLME utilizes constraints of l2 distance and KL distance on representations and probabilities, respectively, to maintain the general capabilities of LLMs.

Key Results

  • Enhanced Monitorability: Across Llama-3.1-8B, Qwen2.5-7B, and Mistral-7B, TELLME achieved a 10.2% average accuracy improvement on representation-based monitors (Self-Similarity and Linear Probing).
  • Efficiency Over Baselines: On linear probing tasks, TELLME outperformed strong SAE and Transcoder baselines by 2.5% on average, despite utilizing 26× fewer trainable parameters (~41M LoRA parameters vs. ~1.09B for the SAE baseline).
  • Active Detoxification: TELLME delivered an absolute average improvement of 8.1 points in safety rates across the BeaverTails and SaladBench benchmarks.
  • Robust Scalability: The method scales seamlessly across parameter sizes (from 7B up to 72B for Qwen2.5) and architectures, showing robust performance gains even on massive Mixture-of-Experts (MoE) multimodal models like Qwen3-VL-235B.
Safety performance improvements across different model scales and architectures

Limitations and Future Work

  • Knowledge Constraints: The effectiveness of representation disentanglement is inherently bounded by the LLM's pre-trained knowledge. Unseen or highly niche behaviors not covered during the fine-tuning process may remain entangled.
  • Post-Training Restriction: Due to computational resource limits, TELLME was only evaluated as a post-training intervention. The authors note that integrating TELLME directly into the pre-training stage could unlock even greater structural transparency and stronger generalization bounds.

Summary updated

Abstract

Large language models (LLMs) are becoming increasingly capable, but the mechanisms of their thinking and decision-making processes remain unclear. Chain-of-thoughts (CoTs) have been commonly utilized to externalize LLMs' thinking, but this strategy fails to accurately reflect LLMs' thinking process. Techniques based on LLMs' hidden representations provide an inner perspective to improve the monitorability of their latent thinking. However, previous methods only try to develop external modules instead of making LLMs themselves easier to monitor. In this paper, we propose a novel method, TELLME, improving the transparency of LLMs and helping monitors identify unsuitable and sensitive behaviors. Furthermore, we showcase the effectiveness of TELLME on detoxification tasks, where LLMs achieve consistent improvement among multimodal test sets, distinct architectures, and varying parameter scales. We further analyze TELLME's improvement on LLMs' generalization ability from both optimal transport theory and empirical perspectives.
BibTeX
@inproceedings{chen2026beyond,
  author = {Guanxu Chen and Jing Shao and Tao Luo and Lijie Hu and Qihao Lin and Dongrui Liu},
  title = {{Beyond External Monitors: Enhancing Transparency of Large Language Models for Easier Monitoring}},
  booktitle = {International Conference on Machine Learning},
  year = {2026},
  url = {https://openreview.net/forum?id=8yrdEj23Hj},
}
Figure