Source author record

Anurag Acharya

Anurag Acharya appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Artificial Intelligence Computation and Language Digital Libraries Machine Learning physics.comp-ph

Catalog footprint

What is connected

6works

5topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

A Cloud-based Multi-Agentic Workflow for Science

As Large Language Models (LLMs) become ubiquitous across various scientific domains, their lack of ability to perform complex tasks like running simulations or to make complex decisions limits their utility. LLM-based agents bridge this gap due to their ability to call external resources and tools and thus are now rapidly gaining popularity. However, coming up with a workflow that can balance the models, cloud providers, and external resources is very challenging, making implementing an agentic system more of a hindrance than a help. In this work, we present a domain-agnostic, model-independent workflow for an agentic framework that can act as a scientific assistant while being run entirely on cloud. Built with a supervisor agent marshaling an array of agents with individual capabilities, our framework brings together straightforward tasks like literature review and data analysis with more complex ones like simulation runs. We describe the framework here in full, including a proof-of-concept system we built to accelerate the study of Catalysts, which is highly important in the field of Chemistry and Material Science. We report the cost to operate and use this framework, including the breakdown of the cost by services use. We also evaluate our system on a custom-curated synthetic benchmark and a popular Chemistry benchmark, and also perform expert validation of the system. The results show that our system is able to route the task to the correct agent 90% of the time and successfully complete the assigned task 97.5% of the time for the synthetic tasks and 91% of the time for real-world tasks, while still achieving better or comparable accuracy to most frontier models, showing that this is a viable framework for other scientific domains to replicate.

preprint2026arXiv

Evaluating Memory Condensation Strategies for Coding Agents in Data-Driven Scientific Discovery

Coding agents accumulate extensive context during long-running tasks, yet fixed context windows force practitioners to choose between truncation and task failure. While numerous memory condensation strategies have been proposed, from simple sliding windows to LLM-generated summaries, no systematic comparison exists to guide strategy selection, especially in scientific discovery tasks. We evaluate eight memory condensation strategies using GPT-4o on sixty DiscoveryBench tasks spanning six scientific domains (480 total evaluations). We find that no condenser significantly alters hypothesis quality, while LLM-based condensers increase token costs by 24-94 percent, and masking tool-call outputs achieves an 8.6 percent net savings. We also observe that the optimal condenser for data-driven scientific discovery varies by scientific domain and task length.

preprint2026arXiv

SCICONVBENCH: Benchmarking LLMs on Multi-Turn Clarification for Task Formulation in Computational Science

Large Language Models (LLMs) are increasingly deployed as scientific AI as- sistants, and a growing body of benchmarks evaluates their capabilities across knowledge retrieval, reasoning, code generation, and tool use. These evaluations, however, typically assume the scientific problem is already well-posed, whereas practical scientific assistance often begins with an ill-posed user request that must be refined through dialogue before any computation, analysis, or experiment can be carried out reliably. We introduce SCICONVBENCH, a benchmark for multi- turn clarification in scientific task formulation across four computational science problem domains: fluid mechanics, solid mechanics, materials science, and par- tial differential equations (PDEs). SCICONVBENCH targets two complementary capabilities: eliciting missing information (disambiguation) and detecting and correcting erroneous requests containing internally contradictory information (in- consistency resolution). Our benchmark pairs a structured task ontology with a rubric-based evaluation framework, enabling systematic measurement of LLM per- formance across three dimensions: clarification behavior, conversational grounding, and final-specification fidelity. Current frontier models perform relatively well on inconsistency resolution, but even the best model resolves only 52.7% of the disambiguation cases in fluid mechanics. We further find that frontier LLMs fre- quently make silent assumptions and perform implicit specification repairs that are not grounded in the conversation with users. SCICONVBENCH establishes a foundation for evaluating the upstream conversational reasoning that a reliable computational science assistant requires. The code and data can be found at https://github.com/csml-rpi/SciConvBench.

preprint2022arXiv

Finding Trolls Under Bridges: Preliminary Work on a Motif Detector

Motifs are distinctive recurring elements found in folklore that have significance as communicative devices in news, literature, press releases, and propaganda. Motifs concisely imply a large constellation of culturally-relevant information, and their broad usage suggests their cognitive importance as touchstones of cultural knowledge, making their detection a worthy step toward culturally-aware natural language processing tasks. Until now, folklorists and others interested in motifs have only extracted motifs from narratives manually. We present a preliminary report on the development of a system for automatically detecting motifs. We briefly describe an annotation effort to produce data for training motif detection, which is on-going. We describe our in-progress architecture in detail, which aims to capture, in part, how people determine whether or not a motif candidate is being used in a motific way. This description includes a test of an off-the-shelf metaphor detector as a feature for motif detection, which achieves a F1 of 0.35 on motifs and a macro-average F1 of 0.21 across four categories which we assign to motif candidates.

preprint2014arXiv

On the Shoulders of Giants: The Growing Impact of Older Articles

In this paper, we examine the evolution of the impact of older scholarly articles. We attempt to answer four questions. First, how often are older articles cited and how has this changed over time. Second, how does the impact of older articles vary across different research fields. Third, is the change in the impact of older articles accelerating or slowing down. Fourth, are these trends different for much older articles. To answer these questions, we studied citations from articles published in 1990-2013. We computed the fraction of citations to older articles from articles published each year as the measure of impact. We considered articles that were published at least 10 years before the citing article as older articles. We computed these numbers for 261 subject categories and 9 broad areas of research. Finally, we repeated the computation for two other definitions of older articles, 15 years and older and 20 years and older. There are three conclusions from our study. First, the impact of older articles has grown substantially over 1990-2013. In 2013, 36% of citations were to articles that are at least 10 years old; this fraction has grown 28% since 1990. The fraction of older citations increased over 1990-2013 for 7 out of 9 broad areas and 231 out of 261 subject categories. Second, the increase over the second half (2002-2013) was double the increase in the first half (1990-2001). Third, the trend of a growing impact of older articles also holds for even older articles. In 2013, 21% of citations were to articles >= 15 years old with an increase of 30% since 1990 and 13% of citations were to articles >= 20 years old with an increase of 36%. Now that finding and reading relevant older articles is about as easy as finding and reading recently published articles, significant advances aren't getting lost on the shelves and are influencing work worldwide for years after.

preprint2014arXiv

Rise of the Rest: The Growing Impact of Non-Elite Journals

In this paper, we examine the evolution of the impact of non-elite journals. We attempt to answer two questions. First, what fraction of the top-cited articles are published in non-elite journals and how has this changed over time. Second, what fraction of the total citations are to non-elite journals and how has this changed over time. We studied citations to articles published in 1995-2013. We computed the 10 most-cited journals and the 1000 most-cited articles each year for all 261 subject categories in Scholar Metrics. We marked the 10 most-cited journals in a category as the elite journals for the category and the rest as non-elite. There are two conclusions from our study. First, the fraction of top-cited articles published in non-elite journals increased steadily over 1995-2013. While the elite journals still publish a substantial fraction of high-impact articles, many more authors of well-regarded papers in diverse research fields are choosing other venues. The number of top-1000 papers published in non-elite journals for the representative subject category went from 149 in 1995 to 245 in 2013, a growth of 64%. Looking at broad research areas, 4 out of 9 areas saw at least one-third of the top-cited articles published in non-elite journals in 2013. For 6 out of 9 areas, the fraction of top-cited papers published in non-elite journals for the representative subject category grew by 45% or more. Second, now that finding and reading relevant articles in non-elite journals is about as easy as finding and reading articles in elite journals, researchers are increasingly building on and citing work published everywhere. Considering citations to all articles, the percentage of citations to articles in non-elite journals went from 27% in 1995 to 47% in 2013. Six out of nine broad areas had at least 50% of citations going to articles published in non-elite journals in 2013.

Institution

Affiliation not imported yet

This author record came from a source that does not expose affiliation metadata. Once the author claims the profile or we enrich the record from another provider, this section will link to the concrete institution.

Topic footprint