arXiv subjects

Vivek Kulkarni

Publications and source records attributed to Vivek Kulkarni.

At least 19 recordsLinked to original sources

arXiv2026

SBCO: Self-Supervised, Verifier-Grounded Harness Optimization For Planning Agents

Self-improving agents seek to reduce the human engineering effort behind AI systems by enabling them to evolve and self-improve their performance over time. Recently, methods like the Darwin G\"odel Machine and the Huxley G\"odel Machine have been proposed which enable open-ended, recursive self-improvement through self-reference where a coding agent edits its own code. Such self-referential self-improvement methods require that the competence required to perform the task coincides or aligns well with the competence required for self-modification which is the case for coding tasks. For domains or tasks, which do not satisfy the alignment needed, self-referential self-improvement is not available. In such cases, it is possible to adapt the above algorithms to other tasks by removing the self-referential aspect or introducing explicit self-modification of a meta-agent -- both computationally expensive, relying on population or self-modification search over many candidate agents. For planning tasks with explicit constraints, we propose a far cheaper alternative. We introduce SBCO (Self-supervised Block Coordinate Optimizer), a verifier-grounded harness optimizer in the same closed-loop, improve-from-experience family as the G\"odel-machine methods, but self-supervised rather than self-referential. Given an agentic harness, SBCO learns a decomposed bank of verifiers and a harness policy via approximate block coordinate ascent, improving the agent's outputs from its own graded feedback---with a fixed meta-agent and no human labels. Across two domains SBCO matches or exceeds a customized self-modifying baseline while using 4-5.5 times less compute budget.

cs.AI↗

arXiv2026

PROGRESS: Coverage-guided RL to Train Search-augmented LLM Agent

Existing search-augmented LLM agents are trained using Reinforcement Learning to boost its reasoning capabilities. However, these approaches primarily rely on outcome-level rewards, which provide little supervision over search behavior and overlook agent's ability to decompose complex queries properly. To mitigate this issue, we propose PROGRESS which utilizes teacher-guided coverage reward to explicitly shape decomposed query generation of the policy model. During training, frozen teacher models are used to decompose complex queries into essential search queries. These essential search queries are utilized to guide the search behavior of the policy model. Integrated into an R1-style training framework, our approach provides lightweight guidance over query decomposition decisions without dense process-level supervision. Experiments show that coverage-guided RL improves overall task performance, highlighting the importance of explicitly supervising search strategies in agentic LLMs.

cs.AI↗

arXiv2026

Search-GRT: Guided Retrieval Training of Search Agents to Optimize for Complex Question Answering

The effective use of search engines by large language models (LLMs) remains a significant challenge, particularly in complex, multi-hop question-answering (MHQA) tasks. These tasks require the model to decompose questions into subqueries, retrieve relevant information, and synthesize answers from multiple sources, often leading to cascading errors due to poor retrieval in early stages. Reinforcement learning (RL) has shown promise in improving LLMs' search capabilities, but it often suffers from sparse rewards during training, hindering the model's ability to learn effectively. To address these challenges, we introduce Guided Retrieval Training (GRT), a novel method that improves the performance of a search agent by restricting the retrieval process during RL training using ground truth information. By focusing on a curated set of relevant documents, GRT provides the model with a stronger learning signal, mitigating the problem of sparse rewards and improving its ability to generate accurate subqueries and synthesize correct answers. Our experimental results demonstrate that GRT achieves consistent performance improvements over existing methods, such as Search-R1, across a wide range of question-answering (QA) tasks. Notably, GRT excels in MHQA tasks, achieving over 40% improvements in performance. Additionally, GRT enhances training efficiency by achieving better QA performance with fewer training steps.

cs.AI↗

arXiv2026

TRUSTMEM: Learning Trustworthy Memory Consolidation for LLM Agents with Long-Term Memory

Large language model (LLM) agents rely on long-term memory to support extended interactions and personalized assistance beyond finite context windows. Existing memory agents actively update external memory through generated write, revise, and delete operations, but these updates may omit important information, corrupt existing memory, or introduce unsupported hallucinated content. Once stored, such errors become persistent system-state failures that can affect future reasoning and generation. In this paper, we propose TrustMem, a framework designed to improve the trustworthiness of memory consolidation. TrustMem relies on a Memory Transition Verifier to evaluate the transition process of memory updates in terms of coverage, preservation, and faithfulness. It further constructs preference pairs among candidate updates under the same memory state, enabling preference-guided reinforcement learning to directly optimize memory updating behaviors. Extensive experiments demonstrate that TrustMem improves both memory utility and reliability: it achieves state-of-the-art results across MemoryAgentBench, HaluMem, and the Mem-alpha validation set, improves HaluMem memory extraction by 12.14 F1 points, and reduces transition-level omission, corruption, and hallucination by 40.1\%, 79.1\%, and 50.0\%, respectively, compared with the strongest baseline for each error type.

cs.AI↗

arXiv2026

Latent Personal Memory: Represent personal memory as dynamic soft prompts

Personalizing large language models (LLMs) requires encoding long-term, user-specific behavioral patterns in a way that is computationally efficient, scalable, and compatible with a frozen base model. We present Latent Personal Memory (LPM), a scalable framework that represents user-specific history as a compact, persistent matrix of N latent slots, that are interpretable. A shared cross-attention projection network maps these slots into dynamic, input-conditioned soft prompts that are prepended to the input of a frozen LLM. We evaluate LPM on PersonaMem v1 and LoCOMO benchmarks across Qwen3-1.7B, 4B, and 8B backbones. Results demonstrate that LPM outperforms LoRA and Prompt Tuning by up to 8.8% and 54.4% in overall accuracy respectively on PersonaMem v1, while reducing KV-cache usage by over 64x. On LoCoMo, LPM matches LoRA accuracy with 120x fewer trainable parameters. We also show that the efficiency of LPM grows with context length and outperforms full-context at 128K context length.

cs.CL↗

arXiv2026

From sunblock to softblock: Analyzing the correlates of neology in published writing and on social media

Living languages are shaped by a host of conflicting internal and external evolutionary pressures. While some of these pressures are universal across languages and cultures, others differ depending on the social and conversational context: language use in newspapers is subject to very different constraints than language use on social media. Prior distributional semantic work on English word emergence (neology) identified two factors correlated with creation of new words by analyzing a corpus consisting primarily of historical published texts (Ryskina et al., 2020, arXiv:2001.07740). Extending this methodology to contextual embeddings in addition to static ones and applying it to a new corpus of Twitter posts, we show that the same findings hold for both domains, though the topic popularity growth factor may contribute less to neology on Twitter than in published writing. We hypothesize that this difference can be explained by the two domains favouring different neologism formation mechanisms.

cs.CL↗

arXiv2025

VOYAGER: A Training Free Approach for Generating Diverse Datasets using LLMs

Large language models (LLMs) are increasingly being used to generate synthetic datasets for the evaluation and training of downstream models. However, prior work has noted that such generated data lacks diversity. In this paper, we propose Voyager, a novel principled approach to generate diverse datasets. Our approach is iterative and directly optimizes a mathematical quantity that optimizes the diversity of the dataset using the machinery of determinantal point processes. Furthermore, our approach is training-free, applicable to closed-source models, and scalable. In addition to providing theoretical justification for the working of our method, we also demonstrate through comprehensive experiments that Voyager significantly outperforms popular baseline approaches by providing a 1.5-3 times improvement in diversity.

cs.CL↗

arXiv2025

APIO: Automatic Prompt Induction and Optimization for Grammatical Error Correction and Text Simplification

Recent advancements in large language models (LLMs) have enabled a wide range of natural language processing (NLP) tasks to be performed through simple prompt-based interactions. Consequently, several approaches have been proposed to engineer prompts that most effectively enable LLMs to perform a given task (e.g., chain-of-thought prompting). In settings with a well-defined metric to optimize model performance, automatic prompt optimization (APO) methods have been developed to refine a seed prompt. Advancing this line of research, we propose APIO, a simple but effective prompt induction and optimization approach for the tasks of Grammatical Error Correction (GEC) and Text Simplification, without relying on manually specified seed prompts. APIO achieves a new state-of-the-art performance for purely LLM-based prompting methods on these tasks. We make our data, code, prompts, and outputs publicly available.

cs.CL↗

arXiv2024

Spivavtor: An Instruction Tuned Ukrainian Text Editing Model

We introduce Spivavtor, a dataset, and instruction-tuned models for text editing focused on the Ukrainian language. Spivavtor is the Ukrainian-focused adaptation of the English-only CoEdIT model. Similar to CoEdIT, Spivavtor performs text editing tasks by following instructions in Ukrainian. This paper describes the details of the Spivavtor-Instruct dataset and Spivavtor models. We evaluate Spivavtor on a variety of text editing tasks in Ukrainian, such as Grammatical Error Correction (GEC), Text Simplification, Coherence, and Paraphrasing, and demonstrate its superior performance on all of them. We publicly release our best-performing models and data as resources to the community to advance further research in this space.

cs.CL↗

arXiv2024

mEdIT: Multilingual Text Editing via Instruction Tuning

We introduce mEdIT, a multi-lingual extension to CoEdIT -- the recent state-of-the-art text editing models for writing assistance. mEdIT models are trained by fine-tuning multi-lingual large, pre-trained language models (LLMs) via instruction tuning. They are designed to take instructions from the user specifying the attributes of the desired text in the form of natural language instructions, such as Grammatik korrigieren (German) or Parafrasee la oración (Spanish). We build mEdIT by curating data from multiple publicly available human-annotated text editing datasets for three text editing tasks (Grammatical Error Correction (GEC), Text Simplification, and Paraphrasing) across diverse languages belonging to six different language families. We detail the design and training of mEdIT models and demonstrate their strong performance on many multi-lingual text editing benchmarks against other multilingual LLMs. We also find that mEdIT generalizes effectively to new languages over multilingual baselines. We publicly release our data, code, and trained models at https://github.com/vipulraheja/medit.

cs.CL↗

arXiv2024

SOCIALITE-LLAMA: An Instruction-Tuned Model for Social Scientific Tasks

Social science NLP tasks, such as emotion or humor detection, are required to capture the semantics along with the implicit pragmatics from text, often with limited amounts of training data. Instruction tuning has been shown to improve the many capabilities of large language models (LLMs) such as commonsense reasoning, reading comprehension, and computer programming. However, little is known about the effectiveness of instruction tuning on the social domain where implicit pragmatic cues are often needed to be captured. We explore the use of instruction tuning for social science NLP tasks and introduce Socialite-Llama -- an open-source, instruction-tuned Llama. On a suite of 20 social science tasks, Socialite-Llama improves upon the performance of Llama as well as matches or improves upon the performance of a state-of-the-art, multi-task finetuned model on a majority of them. Further, Socialite-Llama also leads to improvement on 5 out of 6 related social tasks as compared to Llama, suggesting instruction tuning can lead to generalized social understanding. All resources including our code, model and dataset can be found through bit.ly/socialitellama.

cs.CL↗

arXiv2024

Personalized Text Generation with Fine-Grained Linguistic Control

As the text generation capabilities of large language models become increasingly prominent, recent studies have focused on controlling particular aspects of the generated text to make it more personalized. However, most research on controllable text generation focuses on controlling the content or modeling specific high-level/coarse-grained attributes that reflect authors' writing styles, such as formality, domain, or sentiment. In this paper, we focus on controlling fine-grained attributes spanning multiple linguistic dimensions, such as lexical and syntactic attributes. We introduce a novel benchmark to train generative models and evaluate their ability to generate personalized text based on multiple fine-grained linguistic attributes. We systematically investigate the performance of various large language models on our benchmark and draw insights from the factors that impact their performance. We make our code, data, and pretrained models publicly available.

cs.CL↗

arXiv2023

Writing Assistants Should Model Social Factors of Language

Intelligent writing assistants powered by large language models (LLMs) are more popular today than ever before, but their further widespread adoption is precluded by sub-optimal performance. In this position paper, we argue that a major reason for this sub-optimal performance and adoption is a singular focus on the information content of language while ignoring its social aspects. We analyze the different dimensions of these social factors in the context of writing assistants and propose their incorporation into building smarter, more effective, and truly personalized writing assistants that would enrich the user experience and contribute to increased user adoption.

cs.CL↗

arXiv2022

NTULM: Enriching Social Media Text Representations with Non-Textual Units

On social media, additional context is often present in the form of annotations and meta-data such as the post's author, mentions, Hashtags, and hyperlinks. We refer to these annotations as Non-Textual Units (NTUs). We posit that NTUs provide social context beyond their textual semantics and leveraging these units can enrich social media text representations. In this work we construct an NTU-centric social heterogeneous network to co-embed NTUs. We then principally integrate these NTU embeddings into a large pretrained language model by fine-tuning with these additional units. This adds context to noisy short-text social media. Experiments show that utilizing NTU-augmented text representations significantly outperforms existing text-only baselines by 2-5\% relative points on many downstream tasks highlighting the importance of context to social media NLP. We also highlight that including NTU context into the initial layers of language model alongside text is better than using it after the text embedding is generated. Our work leads to the generation of holistic general purpose social media content embedding.

cs.CL↗

arXiv2022

CTM -- A Model for Large-Scale Multi-View Tweet Topic Classification

Automatically associating social media posts with topics is an important prerequisite for effective search and recommendation on many social media platforms. However, topic classification of such posts is quite challenging because of (a) a large topic space (b) short text with weak topical cues, and (c) multiple topic associations per post. In contrast to most prior work which only focuses on post classification into a small number of topics ($10$-$20$), we consider the task of large-scale topic classification in the context of Twitter where the topic space is $10$ times larger with potentially multiple topic associations per Tweet. We address the challenges above by proposing a novel neural model, CTM that (a) supports a large topic space of $300$ topics and (b) takes a holistic approach to tweet content modeling -- leveraging multi-modal content, author context, and deeper semantic cues in the Tweet. Our method offers an effective way to classify Tweets into topics at scale by yielding superior performance to other approaches (a relative lift of $\mathbf{20}\%$ in median average precision score) and has been successfully deployed in production at Twitter.

cs.CL↗

arXiv2021

Learning, Computing, and Trustworthiness in Intelligent IoT Environments: Performance-Energy Tradeoffs

An Intelligent IoT Environment (iIoTe) is comprised of heterogeneous devices that can collaboratively execute semi-autonomous IoT applications, examples of which include highly automated manufacturing cells or autonomously interacting harvesting machines. Energy efficiency is key in such edge environments, since they are often based on an infrastructure that consists of wireless and battery-run devices, e.g., e-tractors, drones, Automated Guided Vehicle (AGV)s and robots. The total energy consumption draws contributions from multipleiIoTe technologies that enable edge computing and communication, distributed learning, as well as distributed ledgers and smart contracts. This paper provides a state-of-the-art overview of these technologies and illustrates their functionality and performance, with special attention to the tradeoff among resources, latency, privacy and energy consumption. Finally, the paper provides a vision for integrating these enabling technologies in energy-efficient iIoTe and a roadmap to address the open research challenges

cs.DC↗

arXiv2021

LMSOC: An Approach for Socially Sensitive Pretraining

While large-scale pretrained language models have been shown to learn effective linguistic representations for many NLP tasks, there remain many real-world contextual aspects of language that current approaches do not capture. For instance, consider a cloze-test "I enjoyed the ____ game this weekend": the correct answer depends heavily on where the speaker is from, when the utterance occurred, and the speaker's broader social milieu and preferences. Although language depends heavily on the geographical, temporal, and other social contexts of the speaker, these elements have not been incorporated into modern transformer-based language models. We propose a simple but effective approach to incorporate speaker social context into the learned representations of large-scale language models. Our method first learns dense representations of social contexts using graph representation learning algorithms and then primes language model pretraining with these social context representations. We evaluate our approach on geographically-sensitive language-modeling tasks and show a substantial improvement (more than 100% relative lift on MRR) compared to baselines.

cs.CL↗

arXiv2020

TopicBERT for Energy Efficient Document Classification

Prior research notes that BERT's computational cost grows quadratically with sequence length thus leading to longer training times, higher GPU memory constraints and carbon emissions. While recent work seeks to address these scalability issues at pre-training, these issues are also prominent in fine-tuning especially for long sequence tasks like document classification. Our work thus focuses on optimizing the computational cost of fine-tuning for document classification. We achieve this by complementary learning of both topic and language models in a unified framework, named TopicBERT. This significantly reduces the number of self-attention operations - a main performance bottleneck. Consequently, our model achieves a 1.4x ($\sim40\%$) speedup with $\sim40\%$ reduction in $CO_2$ emission while retaining $99.9\%$ performance over 5 datasets.

cs.CL↗