Anurag Mishra

Anurag Mishra

Time series foundation models (TSFMs) are increasingly deployed in high-stakes domains, yet their internal representations remain opaque. We present the first application of sparse autoencoders (SAEs) to a TSFM, training TopK SAEs on activations of Chronos-T5-Large (710M parameters) across six layers. Through 392 single-feature ablation experiments, we establish that every ablated feature produces a positive CRPS degradation, confirming causal relevance. Our analysis reveals a depth-dependent hierarchy: early encoder layers encode low-level frequency features, the mid-encoder concentrates causally critical change-detection features, and the final encoder compresses a rich but less causally important taxonomy of temporal concepts. The most critical features reside in the mid-encoder (max single-feature Delta CRPS = 38.61), not in the semantically richest final encoder layer, where progressive ablation paradoxically improves forecast quality. These findings demonstrate that mechanistic interpretability transfers effectively to TSFMs and that Chronos-T5 relies on abrupt-dynamics detection rather than periodic pattern recognition.

Anurag Mishra

Mechanistic interpretability research seeks to reveal the inner workings of large language models, yet most work focuses on classification or generative tasks rather than summarization. This paper presents an interpretability framework for analyzing how GPT-like models adapt to summarization tasks. We conduct differential analysis between pre-trained and fine-tuned models, quantifying changes in attention patterns and internal activations. By identifying specific layers and attention heads that undergo significant transformation, we locate the "summarization circuit" within the model architecture. Our findings reveal that middle layers (particularly 2, 3, and 5) exhibit the most dramatic changes, with 62% of attention heads showing decreased entropy, indicating a shift toward focused information selection. We demonstrate that targeted LoRA adaptation of these identified circuits achieves significant performance improvement over standard LoRA fine-tuning while requiring fewer training epochs. This work bridges the gap between black-box evaluation and mechanistic understanding, providing insights into how neural networks perform information selection and compression during summarization.

Anurag Mishra

Neural Machine Translation (NMT) models have shown remarkable performance but remain largely opaque in their decision making processes. The interpretability of these models, especially their internal attention mechanisms, is critical for building trust and verifying that these systems behave as intended. In this work, we introduce a systematic framework to quantitatively evaluate the explainability of an NMT model attention patterns by comparing them against statistical alignments and correlating them with standard machine translation quality metrics. We present a set of metrics attention entropy and alignment agreement and validate them on an English-German test subset from WMT14 using a pre trained mT5 model. Our results indicate that sharper attention distributions correlate with improved interpretability but do not always guarantee better translation quality. These findings advance our understanding of NMT explainability and guide future efforts toward building more transparent and reliable machine translation systems.

Anurag Mishra, Ronen Gold, Sanjeev Vijayakumar

Industrial operations have grown exponentially over the last century, driving advancements in energy utilization through vehicles and machinery.This growth has significant environmental implications, necessitating the use of sophisticated technology to monitor and analyze climate data.The surge in industrial activities presents a complex challenge in forecasting its diverse environmental impacts, which vary greatly across different regions.Aim to understand these dynamics more deeply to predict and mitigate the environmental impacts of industrial activities.

Tuhin Sahai, Anurag Mishra, Jose Miguel Pasini et al.

Given a Boolean formula $φ(x)$ in conjunctive normal form (CNF), the density of states counts the number of variable assignments that violate exactly $e$ clauses, for all values of $e$. Thus, the density of states is a histogram of the number of unsatisfied clauses over all possible assignments. This computation generalizes both maximum-satisfiability (MAX-SAT) and model counting problems and not only provides insight into the entire solution space, but also yields a measure for the \emph{hardness} of the problem instance. Consequently, in real-world scenarios, this problem is typically infeasible even when using state-of-the-art algorithms. While finding an exact answer to this problem is a computationally intensive task, we propose a novel approach for estimating density of states based on the concentration of measure inequalities. The methodology results in a quadratic unconstrained binary optimization (QUBO), which is particularly amenable to quantum annealing-based solutions. We present the overall approach and compare results from the D-Wave quantum annealer against the best-known classical algorithms such as the Hamze-de Freitas-Selby (HFS) algorithm and satisfiability modulo theory (SMT) solvers.

Sharad Mehrotra, Shantanu Sharma, Jeffrey D. Ullman et al.

Despite extensive research on cryptography, secure and efficient query processing over outsourced data remains an open challenge. This paper continues along the emerging trend in secure data processing that recognizes that the entire dataset may not be sensitive, and hence, non-sensitivity of data can be exploited to overcome limitations of existing encryption-based approaches. We propose a new secure approach, entitled query binning (QB) that allows non-sensitive parts of the data to be outsourced in clear-text while guaranteeing that no information is leaked by the joint processing of non-sensitive data (in clear-text) and sensitive data (in encrypted form). QB maps a query to a set of queries over the sensitive and non-sensitive data in a way that no leakage will occur due to the joint processing over sensitive and non-sensitive data. Interestingly, in addition to improve performance, we show that QB actually strengthens the security of the underlying cryptographic technique by preventing size, frequency-count, and workload-skew attacks.