arXiv subjects

Roshan Klein-Seetharaman

Publications and source records attributed to Roshan Klein-Seetharaman.

4 recordsLinked to original sources

arXiv2026

Lomekwi: Resource-Bounded Tool Discovery in LLM Agents

Existing tool-use benchmarks report a single success rate for complex, multistep tasks. Inspired by ideas from cognitive science, we distinguish tool use from tool discovery and decompose the latter into curiosity (the model's ability to discover the parts needed to build the tool), recognition (the model's ability to discover the process of creating the tool), and efficiency (the model's use of the tool after creation). We show that this framework can be applied to existing discovery tasks, such as Voyager. In addition, we provide evidence that recognition inversely scales with model size, and we introduce and analyze a class of combinatorial games that demonstrates this. We further observe inverse scaling in a separate environment designed to emulate real-world tasks.

cs.AI↗

arXiv2025

Beyond URDF: The Universal Robot Description Directory for Shared, Extensible, and Standardized Robot Models

Robots are typically described in software by specification files (e.g., URDF, SDF, MJCF, USD) that encode only basic kinematic, dynamic, and geometric information. As a result, downstream applications such as simulation, planning, and control must repeatedly re-derive richer data, leading to redundant computations, fragmented implementations, and limited standardization. In this work, we introduce the Universal Robot Description Directory (URDD), a modular representation that organizes derived robot information into structured, easy-to-parse JSON and YAML modules. Our open-source toolkit automatically generates URDDs from URDFs, with a Rust implementation supporting Bevy-based visualization. Additionally, we provide a JavaScript/Three.js viewer for web-based inspection of URDDs. Experiments on multiple robot platforms show that URDDs can be generated efficiently, encapsulate substantially richer information than standard specification files, and directly enable the construction of core robotics subroutines. URDD provides a unified, extensible resource for reducing redundancy and establishing shared standards across robotics frameworks. We conclude with a discussion on the limitations and implications of our work.

cs.RO↗

arXiv2023

Optimal connectivity results for spheres in the curve graph of low and medium complexity surfaces

Answering a question of Wright, we show that spheres of any radius are always connected in the curve graph of surfaces $Σ_{2,0}, Σ_{1,3},$ and $Σ_{0,6}$, and the union of two consecutive spheres is always connected for $Σ_{0, 5}$ and $Σ_{1,2}$. We also classify the connected components of spheres of radius 2 in the curve graph of $Σ_{0, 5}$ and $Σ_{1,2}$.

math.GT↗

arXiv2020

The arrow of time (second law) as a randomness-driven emergent property of large systems

The arrow of time is an irreversible phenomenon for a system of particles undergoing reversible dynamics. Since the time of Boltzmann to this day, the arrow of time has led to debate and research. However, the enormous growth of nanotechnology and associated experimental techniques has brought the arrow of time at the forefront because of its practical implications. Using simulations of one-dimensional diffusion of a system of particles, we show that the arrow of time is an emergent property of a large system. We show that the recurrence time for a system of particles to return to its original configuration grows rapidly as the number of particles grows. Based on the simulations, we have provided the expressions for recurrence times for classical particles, Fermions, and Bosons. A system of Bosons has the shortest recurrence time, whereas a system of classical particles has the longest recurrence time. The underlying distribution around the mean recurrence time is Poisson-distributed for Bosons and Gaussian-distributed for Fermions and classical particles. The probabilistic approach to encode dynamics enables testing processes other than diffusion and quantify their effects on the recurrence time.

cond-mat.stat-mech↗