arXiv subjects

Souymodip Chakraborty

Publications and source records attributed to Souymodip Chakraborty.

5 recordsLinked to original sources

arXiv2026

VectorArk: Learning Practical Image Vectorization with Rounded Polygon Representation

Recent vision-language model (VLM)-based approaches have achieved impressive results on image vectorization tasks. However, they are typically evaluated on synthetic benchmarks, where clean SVGs are rasterized at high resolution and then re-vectorized. As a result, these methods generalize poorly to real-world scenarios, such as images with unknown rasterization methods or those generated by text-to-image models. We introduce VectorArk, a new VLM-based model designed for robust and practical image vectorization. VectorArk employs a novel rounded polygon representation that simplifies the learning process while naturally producing smooth, visually appealing primitives. We also propose a degradation model that enhances robustness across diverse and imperfect inputs. Our experiments show that, in contrast to previous methods, VectorArk achieves superior geometric completeness and artifact suppression across multiple datasets, with comprehensive ablations validating the contribution of each component.

cs.CV↗

arXiv2025

Taming Identity Consistency and Prompt Diversity in Diffusion Models via Latent Concatenation and Masked Conditional Flow Matching

Subject-driven image generation aims to synthesize novel depictions of a specific subject across diverse contexts while preserving its core identity features. Achieving both strong identity consistency and high prompt diversity presents a fundamental trade-off. We propose a LoRA fine-tuned diffusion model employing a latent concatenation strategy, which jointly processes reference and target images, combined with a masked Conditional Flow Matching (CFM) objective. This approach enables robust identity preservation without architectural modifications. To facilitate large-scale training, we introduce a two-stage Distilled Data Curation Framework: the first stage leverages data restoration and VLM-based filtering to create a compact, high-quality seed dataset from diverse sources; the second stage utilizes these curated examples for parameter-efficient fine-tuning, thus scaling the generation capability across various subjects and contexts. Finally, for filtering and quality assessment, we present CHARIS, a fine-grained evaluation framework that performs attribute-level comparisons along five key axes: identity consistency, prompt adherence, region-wise color fidelity, visual quality, and transformation diversity.

cs.CV↗

arXiv2025

Beyond the Pixels: VLM-based Evaluation of Identity Preservation in Reference-Guided Synthesis

Evaluating identity preservation in generative models remains a critical yet unresolved challenge. Existing metrics rely on global embeddings or coarse VLM prompting, failing to capture fine-grained identity changes and providing limited diagnostic insight. We introduce Beyond the Pixels, a hierarchical evaluation framework that decomposes identity assessment into feature-level transformations. Our approach guides VLMs through structured reasoning by (1) hierarchically decomposing subjects into (type, style) -> attribute -> feature decision tree, and (2) prompting for concrete transformations rather than abstract similarity scores. This decomposition grounds VLM analysis in verifiable visual evidence, reducing hallucinations and improving consistency. We validate our framework across four state-of-the-art generative models, demonstrating strong alignment with human judgments in measuring identity consistency. Additionally, we introduce a new benchmark specifically designed to stress-test generative models. It comprises 1,078 image-prompt pairs spanning diverse subject types, including underrepresented categories such as anthropomorphic and animated characters, and captures an average of six to seven transformation axes per prompt.

cs.CV↗

arXiv2015

On the Hardness of PCTL Satisfiability

This paper shows that the satisfiability problem for probabilistic CTL (PCTL, for short) is undecidable. By a reduction from $1\frac{1}{2}$-player games with PCTL winning objectives, we establish that the PCTL satisfiability problem is $Σ_1^1$-hard. We present an exponential-time algorithm for the satisfiability of a bounded, negation-closed fragment of PCTL, and show that the satisfiability problem for this fragment is EXPTIME-hard.

cs.LO↗

arXiv2014

Parametric LTL on Markov Chains

This paper is concerned with the verification of finite Markov chains against parametrized LTL (pLTL) formulas. In pLTL, the until-modality is equipped with a bound that contains variables; e.g., $\Diamond_{\le x}\ φ$ asserts that $φ$ holds within $x$ time steps, where $x$ is a variable on natural numbers. The central problem studied in this paper is to determine the set of parameter valuations $V_{\prec p} (φ)$ for which the probability to satisfy pLTL-formula $φ$ in a Markov chain meets a given threshold $\prec p$, where $\prec$ is a comparison on reals and $p$ a probability. As for pLTL determining the emptiness of $V_{> 0}(φ)$ is undecidable, we consider several logic fragments. We consider parametric reachability properties, a sub-logic of pLTL restricted to next and $\Diamond_{\le x}$, parametric Büchi properties and finally, a maximal subclass of pLTL for which emptiness of $V_{> 0}(φ)$ is decidable.

cs.LO↗