arXiv subjects

Aditya Choudhary

Publications and source records attributed to Aditya Choudhary.

10 recordsLinked to original sources

arXiv2026

When Fine-Tuning Fails and when it Generalises: Role of Data Diversity and Mixed Training in LLM-based TTS

Large language models are increasingly adopted as semantic backbones for neural text-to-speech systems. However, frozen LLM representations are insufficient for modeling speaker specific acoustic and perceptual characteristics. Our experiments involving fine tuning of the Language Model backbone of TTS show promise in improving the voice consistency and Signal to Noise ratio SNR in voice cloning task. Across multiple speakers LoRA finetuning consistently outperforms the non-finetuned base Qwen-0.5B model across three complementary dimensions of speech quality. First, perceptual quality improves significantly with DNS-MOS gains of up to 0.42 points for speakers whose training data exhibits sufficient acoustic variability. Second, speaker fidelity improves for all evaluated speakers with consistent increases in voice similarity indicating that LoRA effectively adapts speaker identity representations without degrading linguistic modeling. Third, signal level quality improves in most cases with signal to noise ratio increasing by as much as 34 percent. Crucially these improvements are strongly governed by the characteristics of the training data. Speakers with high variability in acoustic energy and perceptual quality achieve simultaneous gains in DNS-MOS voice similarity and SNR. Overall this work establishes that LoRA finetuning is not merely a parameter efficient optimization technique but an effective mechanism for better speaker level adaptation in compact LLM-based TTS systems. When supported by sufficiently diverse training data LoRA adapted Qwen-0.5B consistently surpasses its frozen base model in perceptual quality speaker similarity with low latency using GGUF model hosted in quantized form.

cs.SD↗

arXiv2026

MM-tau-p$^2$: Persona-Adaptive Prompting for Robust Multi-Modal Agent Evaluation in Dual-Control Settings

Current evaluation frameworks and benchmarks for LLM powered agents focus on text chat driven agents, these frameworks do not expose the persona of user to the agent, thus operating in a user agnostic environment. Importantly, in customer experience management domain, the agent's behaviour evolves as the agent learns about user personality. With proliferation of real time TTS and multi-modal language models, LLM based agents are gradually going to become multi-modal. Towards this, we propose the MM-tau-p$^2$ benchmark with metrics for evaluating the robustness of multi-modal agents in dual control setting with and without persona adaption of user, while also taking user inputs in the planning process to resolve a user query. In particular, our work shows that even with state of-the-art frontier LLMs like GPT-5, GPT 4.1, there are additional considerations measured using metrics viz. multi-modal robustness, turn overhead while introducing multi-modality into LLM based agents. Overall, MM-tau-p$^2$ builds on our prior work FOCAL and provides a holistic way of evaluating multi-modal agents in an automated way by introducing 12 novel metrics. We also provide estimates of these metrics on the telecom and retail domains by using the LLM-as-judge approach using carefully crafted prompts with well defined rubrics for evaluating each conversation.

cs.ET↗

arXiv2026

Boltzmann Reinforcement Learning for Noise resilience in Analog Ising Machines

Analog Ising machines (AIMs) have emerged as a promising paradigm for combinatorial optimization, utilizing physical dynamics to solve Ising problems with high energy efficiency. However, the performance of traditional optimization and sampling algorithms on these platforms is often limited by inherent measurement noise. We introduce BRAIN (Boltzmann Reinforcement for Analog Ising Networks), a distribution learning framework that utilizes variational reinforcement learning to approximate the Boltzmann distribution. By shifting from state-by-state sampling to aggregating information across multiple noisy measurements, BRAIN is resilient to Gaussian noise characteristic of AIMs. We evaluate BRAIN across diverse combinatorial topologies, including the Curie-Weiss and 2D nearest-neighbor Ising systems. We find that under realistic 3\% Gaussian measurement noise, BRAIN maintains 98\% ground state fidelity, whereas Markov Chain Monte Carlo (MCMC) methods degrade to 51\% fidelity. Furthermore, BRAIN reaches the MCMC-equivalent solution up to 192x faster under these conditions. BRAIN exhibits $\mathcal{O}(N^{1.55})$ scaling up to 65,536 spins and maintains robustness against severe measurement uncertainty up to 40\%. Beyond ground state optimization, BRAIN accurately captures thermodynamic phase transitions and metastable states, providing a scalable and noise-resilient method for utilizing analog computing architectures in complex optimizations.

cs.LG↗

arXiv2026

FOCAL: A Novel Benchmarking Technique for Multi-modal Agents

With the recent advancements in reasoning capabilities, tool calling using MCP servers and Audio Language Models (ALMs), development and integration of multi-modal agents (with voice and text support) has come to the industry forefront. Cascading pipelines for voice agents still play a central role in the industry owing to their superior reasoning capabilities facilitated by LLMs. Although, cascading pipelines often present error propagation through the pipeline. We propose a framework, FOCAL to benchmark end-to-end reasoning, component-wise error propagation and error analysis for automated as well as human-assisted testing of multi-modal agents (voice to voice + text input). We also share two novel metrics viz. Reasoning and Semantic scores to evaluate efficacy of the agent in having meaningful conversations in voice mode.

cs.SD↗

arXiv2025

State-of-the-art Small Language Coder Model: Mify-Coder

We present Mify-Coder, a 2.5B-parameter code model trained on 4.2T tokens using a compute-optimal strategy built on the Mify-2.5B foundation model. Mify-Coder achieves comparable accuracy and safety while significantly outperforming much larger baseline models on standard coding and function-calling benchmarks, demonstrating that compact models can match frontier-grade models in code generation and agent-driven workflows. Our training pipeline combines high-quality curated sources with synthetic data generated through agentically designed prompts, refined iteratively using enterprise-grade evaluation datasets. LLM-based quality filtering further enhances data density, enabling frugal yet effective training. Through disciplined exploration of CPT-SFT objectives, data mixtures, and sampling dynamics, we deliver frontier-grade code intelligence within a single continuous training trajectory. Empirical evidence shows that principled data and compute discipline allow smaller models to achieve competitive accuracy, efficiency, and safety compliance. Quantized variants of Mify-Coder enable deployment on standard desktop environments without requiring specialized hardware.

cs.SE↗

arXiv2025

i-LAVA: Insights on Low Latency Voice-2-Voice Architecture for Agents

We experiment with a low-latency, end-to-end voice-to-voice communication model to optimize it for real-time conversational applications. By analyzing components essential to voice to voice (V-2-V) system viz. automatic speech recognition (ASR), text-to-speech (TTS), and dialog management, our work analyzes how to reduce processing time while maintaining high-quality interactions to identify the levers for optimizing V-2-V system. Our work identifies that TTS component which generates life-like voice, full of emotions including natural pauses and exclamations has highest impact on Real time factor (RTF). The experimented V-2-V architecture utilizes CSM1b has the capability to understand tone as well as context of conversation by ingesting both audio and text of prior exchanges to generate contextually accurate speech. We explored optimization of Residual Vector Quantization (RVQ) iterations by the TTS decoder which come at a cost of decrease in the quality of voice generated. Our experimental evaluations also demonstrate that for V-2-V implementations based on CSM most important optimizations can be brought by reducing the number of RVQ Iterations along with the codebooks used in Mimi.

cs.SD↗

arXiv2025

Efficient Malware Detection with Optimized Learning on High-Dimensional Features

Malware detection using machine learning requires feature extraction from binary files, as models cannot process raw binaries directly. A common approach involves using LIEF for raw feature extraction and the EMBER vectorizer to generate 2381-dimensional feature vectors. However, the high dimensionality of these features introduces significant computational challenges. This study addresses these challenges by applying two dimensionality reduction techniques: XGBoost-based feature selection and Principal Component Analysis (PCA). We evaluate three reduced feature dimensions (128, 256, and 384), which correspond to approximately 5.4%, 10.8%, and 16.1% of the original 2381 features, across four models-XGBoost, LightGBM, Extra Trees, and Random Forest-using a unified training, validation, and testing split formed from the EMBER-2018, ERMDS, and BODMAS datasets. This approach ensures generalization and avoids dataset bias. Experimental results show that LightGBM trained on the 384-dimensional feature set after XGBoost feature selection achieves the highest accuracy of 97.52% on the unified dataset, providing an optimal balance between computational efficiency and detection performance. The best model, trained in 61 minutes using 30 GB of RAM and 19.5 GB of disk space, generalizes effectively to completely unseen datasets, maintaining 95.31% accuracy on TRITIUM and 93.98% accuracy on INFERNO. These findings present a scalable, compute-efficient approach for malware detection without compromising accuracy.

cs.CR↗

arXiv2025

An Efficient Method for Accurate Pose Estimation and Error Correction of Cuboidal Objects

The proposed system outlined in this paper is a solution to a use case that requires the autonomous picking of cuboidal objects from an organized or unorganized pile with high precision. This paper presents an efficient method for precise pose estimation of cuboid-shaped objects, which aims to reduce errors in target pose in a time-efficient manner. Typical pose estimation methods like global point cloud registrations are prone to minor pose errors for which local registration algorithms are generally used to improve pose accuracy. However, due to the execution time overhead and uncertainty in the error of the final achieved pose, an alternate, linear time approach is proposed for pose error estimation and correction. This paper presents an overview of the solution followed by a detailed description of individual modules of the proposed algorithm.

cs.CV↗

arXiv2025

Enhancing Plagiarism Detection in Marathi with a Weighted Ensemble of TF-IDF and BERT Embeddings for Low-Resource Language Processing

Plagiarism involves using another person's work or concepts without proper attribution, presenting them as original creations. With the growing amount of data communicated in regional languages such as Marathi -- one of India's regional languages -- it is crucial to design robust plagiarism detection systems tailored for low-resource languages. Language models like Bidirectional Encoder Representations from Transformers (BERT) have demonstrated exceptional capability in text representation and feature extraction, making them essential tools for semantic analysis and plagiarism detection. However, the application of BERT for low-resource languages remains under-explored, particularly in the context of plagiarism detection. This paper presents a method to enhance the accuracy of plagiarism detection for Marathi texts using BERT sentence embeddings in conjunction with Term Frequency-Inverse Document Frequency (TF-IDF) feature representation. This approach effectively captures statistical, semantic, and syntactic aspects of text features through a weighted voting ensemble of machine learning models.

cs.CL↗

arXiv2018

Quantum Mechanical Process of Carbonate Complex Formation and Large Scale Anisotropy in the Adsorption Energy of CO$_2$ on Anatase TiO$_2$ (001) Surface

Adsorption of CO$_2$ on a semiconductor surface is a prerequisite for its photocatalytic reduction. Owing to superior photocorrosion resistance, nontoxicity and suitable band edge positions, TiO$_2$ is considered to be the most efficient photocatalyst for facilitating redox reactions. However, due to the absence of adequate understanding of the mechanism of adsorption, the CO$_2$ conversion efficiency on TiO$_2$ surfaces has not been maximized. While anatase TiO$_2$ (101) is the most stable facet, the (001) surface is more reactive and it has been experimentally shown that the stability can be reversed and a larger percentage (up to ~ 89%) of the (001) facet can be synthesized in the presence fluorine ions. Therefore, through density functional calculations we have investigated the CO$_2$ adsorption on TiO$_2$ (001) surface. We have developed a three-state quantum-mechanical model that explains the mechanism of chemisorption, leading to the formation of a tridentate carbonate complex. The electronic structure analysis reveals that the CO$_2$-TiO$_2$ interaction at the surface is uniaxial and long ranged, which gives rise to anisotropy in binding energy (BE). It negates the widely perceived one-to-one correspondence between coverage and BE and infers that the spatial distribution of CO$_2$ primarily determines the BE. A conceptual experiment is devised where the CO$_2$ concentration and flow direction can be controlled to tune the BE within a large window of ~1.5 eV. The experiment also reveals that a maximum of 50% coverage can be achieved for chemisorption. In the presence of water, the activated carbonate complex forms a bicarbonate complex by overcoming a potential barrier of ~0.9 eV.

cond-mat.mtrl-sci↗