# TARDIS: Mitigating Temporal Misalignment via Representation Steering

arXiv: https://arxiv.org/abs/2503.18693 · HTML: https://mlsanity.com/paper/2503.18693/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Activation steering methods, one of 13 directions in Mechanistic interpretability.
- Builds on 12 of its 24 references that are in the map.
- First posted on arXiv 24 Mar 2025.

## Placed in

- Mechanistic interpretability / [Activation steering methods](https://mlsanity.com/mechinterp/activation-steering-methods/)

## Builds on (most cited first, 12 of 12)

- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [Representation Engineering: A Top-Down Approach to AI Transparency](https://mlsanity.com/paper/2310.01405/) (2023)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering](https://mlsanity.com/paper/2311.06668/) (2023)
- [A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO and Toxicity](https://mlsanity.com/paper/2401.01967/) (2024)
- [Linear Representations of Sentiment in Large Language Models](https://mlsanity.com/paper/2310.15154/) (2023)
- [Time Waits for No One! Analysis and Challenges of Temporal Misalignment](https://mlsanity.com/paper/2111.07408/) (2021)
- [Temporal Adaptation of BERT and Performance on Downstream Document Classification: Insights from Social Media](https://mlsanity.com/paper/2104.08116/) (2021)
- [Time is Encoded in the Weights of Finetuned Language Models](https://mlsanity.com/paper/2312.13401/) (2023)
- [Discovering Bias in Latent Space: An Unsupervised Debiasing Approach](https://mlsanity.com/paper/2406.03631/) (2024)
- [Time-Aware Representation Learning for Time-Sensitive Question Answering](https://mlsanity.com/paper/2310.12585/) (2023)

## Built on by (most cited first, 1 of 1)

- [Temporal Preference Concepts and their Functions in a Large Language Model](https://mlsanity.com/paper/2606.05194/) (2026)

## Abstract (arXiv)

Language models often struggle with temporal misalignment, performance degradation caused by shifts in the temporal distribution of data. Continuously updating models to avoid degradation is expensive. Can models be adapted without updating model weights? We present TARDIS, an unsupervised representation editing method that addresses this challenge. TARDIS extracts steering vectors from unlabeled data and adjusts the model's representations to better align with the target time period's distribution. Our experiments reveal that TARDIS enhances downstream task performance without the need for fine-tuning, can mitigate temporal misalignment even when exact target time period data is unavailable, and remains efficient even when the temporal information of the target data points is unknown at inference time.