Controls on Omanshu Thapliyal

Contraction can Help Safety Classifiers keep their Promise

Wed, 12 Aug 2026 01:02:49 -0400

The code for this project can be found at: https://github.com/omanshuthapliyal/contraction_constrained-safety-head. And the paper can be found at: https://arxiv.org/pdf/2610.02853.

A safety classifier can tell us that a prompt looks harmful. That is useful. But it is perhaps not the question we should stop at. Suppose a classifier sees a prompt and says that it is safe. Now suppose we change the prompt slightly. Perhaps we substitute a few words. Perhaps we phrase the same request differently. Perhaps, more abstractly, we nudge the token embeddings of the prompt by a very small amount. Would the classifier still say safe?

Can LLM Finetuning be Stateful? Small states for Large Models - Part 2

Fri, 07 Aug 2026 11:57:21 -0700

The code for this project can be found at: https://github.com/omanshuthapliyal/HRM-adapter. And the paper can be found at: https://arxiv.org/pdf/2606.26290.

In the previous post, I introduced the basic idea behind a stateful adapter. Instead of asking a small trainable module to only modify the current transformer representation, we allow it to maintain a state: $$ s_t = \bar{A}s_{t-1} + \bar{B}h_t, $$ and use that state to generate a residual correction, $$ r_t = Cs_t. $$ The important unresolved question was: if this adapter gets a small amount of memory, what should it remember? The answer I want to explore is the same one that appeared in the discussion of Hankel singular values:

Can LLM Finetuning be Stateful? Small states for Large Models - Part 1

Fri, 31 Jul 2026 11:57:21 -0700

In the last post I talked about Hankel operators of simple linear dynamical systems. Such operators, and either eigenvalues, “can be used as a proxy for how much memory a task actually requires in a state-space model”. And later we tied their input-output relations empirically tie the memory requirements of a post-trained state space model (SSM) to its Hankel singular value (HSV) rates. Further, HSV-based balanced truncation can offer us to selectively choose ‘some’ eigenvalues to keep that involve higher input-to-output signal strength over time. And so, we used Hankel singular values to decide which states of a dynamical system are worth keeping. However, a language model also processes a sequence of tokens over time. We saw task-dependent HSV decay rates for various tasks for different SSM-based models. Similarly, could we give the model a small state, one deliberately built around that same notion of “worth keeping”, without retraining the entire model?

State Space Models: How Control Theory can echo Machine Learning

Mon, 06 Oct 2025 18:57:21 -0700

I wrote in a previous post how transformers are kind of like system identification methods applied to some sequence in a state space. In this post we try to understand if the other way round is true.

We already saw previously ¹ how predicting words/tokens is a sequence to sequence mapping problem, which is the subject of a lot of prediction, estimation, and controls problems. In fact, we leveraged a transformer’s sequence mapping ability and applied it to a controls problem of trajectory prediction for a simple 2D car model ¹. Can we do the other way round? Turns out: yes we can!

How Transformers Echo Control Theory

Sat, 27 Sep 2025 11:57:21 -0700

Transformer models have captivated a lot of AI research in most of the past decade, and in this post my goal is to make them seem more interesting to controls people, who have much to contribute to the field. I will be making gross generalizations to draw such entry points. At least, I found transformer models more interesting after finding a few parallels in dynamical systems & controls ¹. This post attempts to explore those parallels.

Technology Readiness Levels of AI methods

Tue, 15 Oct 2024 11:07:46 -0700

Recent news of the SpaceX catching its Starship Super Heavy booster¹ is quite discussed and marveled upon in the media, and rightly so. Executing what was done can perhaps be explained as trying to catch a falling stick, then balancing it vertically, all while the stick is over 200 ft tall, weighs over 150 tonnes, spews fire, and costs over tens of millions of dollars. One cannot but feel a little giddy as a controls engineering enthusiast over this massive success. It makes me think why we are closer to automation in certain domains/problems than others? Why, for example, is your vacuum cleaning robot seemingly better at navigating in a cluttered environment than your possibly self-driving car? Why is the tremendous progress in machine learning & AI not permeating outside of our web lives as much as we expected in the early 2000’s?

Why I am studying this: Data Driven Safety

Wed, 16 Mar 2022 12:32:31 -0400

This is my attempt to try to motivate and justify why I am studying what I study, and to try to justify the title of my dissertation.

While most emphasis on safety has been provided from a model-driven perspective, data-driven approaches present a novel semi model-driven method by employing model-discovery. As data becomes more inexpensive, and system capabilities highly nonlinear, and often subject to parameter variations during mission, data-driven approaches are utilized more and more in controls engineering. Further, online and real-time applications increasingly require unknown models and nonlinear interactions to be learned for control synthesis. To this end, more established fields of system identification are increasingly utilizing data-driven models by using neural networks, autoregressive models, and other machine learning techniques. Due to the expressive nature of such machine learning techniques, they are utilized to learn models of unknown dynamical systems from observed data sequences. In such approaches, a `discovered model’ is arrived at using data, and employed to perform control design. Even though purely model-driven are slowly getting harder to apply to more recent, realistic, and complex scenarios, they provide a strong basis to study extensions of similar approaches to data-driven techniques for estimation and control. As a result, I attempt to utilize model-driven techniques and extend them to study safety and security properties of cyberphysical systems using data. Therefore, my primary aim is to study data-driven approaches to safety, and data-driven vulnerabilities against cyberattacks, of complex control systems.

The Industry-Academia split in Control Systems

Mon, 16 Sep 2019 12:32:31 -0400

I often wonder why the “cutting edge” research that we read about in the foremost scientific journals & conferences in Control Systems are never applied in the industry. Because whenever we read/write papers, Section-I always tries to ground itself to industry and/or practical applications. However, it’s pretty well known that a majority of the controllers used in the industry are still tuned PIDs. Even though sub-fields of modern control – adaptive, and robust control are very mature and pretty old in their own rights, the real world’s inertia against using modern control techniques bears substantial weight.