Searcharxiv⌕ Search

arXiv subjects

Ivan Nasonov

Publications and source records attributed to Ivan Nasonov.

7 recordsLinked to original sources
arXiv2026

EmoMed: An Emotionally-Aware Agent for Multimodal Medical Support with Real-Time Information Retrieval

We present EmoMed - a multimodal medical consultation agent that adapts its responses based on users' emotional states while maintaining clinical accuracy. The system processes text and medical images, detects affect indicators (anxiety, confusion, urgency) from user input, and adjusts response tone, structure, and detail level accordingly. To ensure factual reliability, the agent grounds clinical information through a dual retrieval mechanism: web-based fact-checking and an API-connected, continuously updated medical knowledge base. We evaluate our approach across seven state-of-the-art language models (GPT-4/5, Qwen3, Llama 4, Gemini 2.5, Grok4, Claude3) using comprehensive metrics including LLM-as-judge assessments, MedQA style accuracy tests, BERT Score, safety/helpfulness ratings, and multimodal medical benchmarks. The results demonstrate that emotionally adaptive responses consistently outperform neutral baseline across evaluation dimensions, without compromising clinical accuracy. A controlled user study validated these findings, with participants reporting improved perceived empathy and communication clarity, while maintaining trust in factual accuracy. Source code: https://github.com/NasonovIvan/EmoMed-Agent

cs.AI↗
arXiv2026

ReDAct: Uncertainty-Aware Deferral for LLM Agents

Recently, LLM-based agents have become increasingly popular across many applications, including complex sequential decision-making problems. However, they inherit the tendency of LLMs to hallucinate, leading to incorrect decisions. In sequential settings, even a single mistake can irreversibly degrade the trajectory, making hallucinations an even bigger problem. Although larger LLMs hallucinate less, they incur a significantly higher per-token cost. In this paper, we address this tradeoff by proposing ReDAct (Reason-Defer-Act). In ReDAct, an agent is equipped with two LLMs: a small, cheap model used by default, and a large, more reliable but expensive model. When the predictive uncertainty of the small model exceeds a calibrated threshold, the decision is deferred to the large model. We evaluate our approach in text-based embodied environments such as ALFWorld and MiniGrid and show that deferring only about 15% of decisions to the large model can match the quality of using it exclusively, while significantly reducing inference costs.

cs.CL↗
arXiv2026

Affine transverse foliations in sphere bundles

Let~$S^{n-1}\rightarrow E \rightarrow M^n$ be an oriented sphere bundle supporting a smooth affine transverse foliation. We give an upper bound for the Euler number of the bundle. We also give a new and elementary proof of the following fact: if the fundamental group $\pi_1(M^n)$ is amenable, then the Euler number of the bundle vanishes.

math.GT↗
arXiv2024

Each generic polytope in $\mathbb{R}^3$ has a point with ten normals to the boundary

It is conjectured since long that each smooth convex body $\mathbf{P}\subset \mathbb{R}^n$ has a point in its interior which belongs to at least $2n$ normals from different points on the boundary of $\mathbf{P}$. The conjecture is proven for $n=2,3,4$. We treat the same problem for convex polytopes in $\mathbb{R}^3$ and prove that each generic polytope has a point in its interior with at least $10$ normals to the boundary. This bound is exact: there exists a tetrahedron with no more than $10$ normals emanating from a point in its interior. The proof is based on piecewise linear analog of Morse theory, analysis of bifurcations, and some combinatorial tricks.

math.MG↗
arXiv2024

Concurrent normals problem for convex polytopes and Euclidean distance degree

It is conjectured since long that for any convex body $P\subset \mathbb{R}^n$ there exists a point in its interior which belongs to at least $2n$ normals from different points on the boundary of $P$. The conjecture is known to be true for $n=2,3,4$. We treat the same problem for convex polytopes in $\mathbb{R}^3$. It turns out that the PL concurrent normals problem differs a lot from the smooth one. One almost immediately proves that a convex polytope in $\mathbb{R}^3$ has $8$ normals to its boundary emanating from some point in its interior. Moreover, we conjecture that each simple polytope in $\mathbb{R}^3$ has a point in its interior with $10$ normals to the boundary. We confirm the conjecture for all tetrahedra and triangular prisms and give a sufficient condition for a simple polytope to have a point with $10$ normals. Other related topics (average number of normals, minimal number of normals from an interior point, other dimensions) are discussed.

math.MG↗
arXiv2023

Neural Network Algorithm for Intercepting Targets Moving Along Known Trajectories by a Dubins' Car

The task of intercepting a target moving along a rectilinear or circular trajectory by a Dubins' car is formulated as a time-optimal control problem with an arbitrary direction of the car's velocity at the interception moment. To solve this problem and to synthesize interception trajectories, neural network methods of unsupervised learning based on the Deep Deterministic Policy Gradient algorithm are used. The analysis of the obtained control laws and interception trajectories in comparison with the analytical solutions of the interception problem is performed. The mathematical modeling for the parameters of the target movement that the neural network had not seen before during training is carried out. Model experiments are conducted to test the stability of the neural solution. The effectiveness of using neural network methods for the synthesis of interception trajectories for given classes of target movements is shown.

math.OC↗