arXiv subjects

Robert Nowak

Publications and source records attributed to Robert Nowak.

At least 19 recordsLinked to original sources

arXiv2026

Ludi${}_{\scriptscriptstyle 0.1}$: An Agentic System for Socially Intelligent Robots

Robot foundation models have substantially advanced perception and control, but natural human-robot collaboration requires more than executing isolated commands. A robot must recognize ambiguity, maintain context across turns, communicate its intentions, and revise ongoing behavior as the user's intent changes. We present $\scriptstyle\mathsf{Ludi}_{\scriptscriptstyle 0.1}$, an agentic system for socially intelligent robots that integrates interactive speech, multimodal reasoning, memory, navigation, and learned manipulation. Its decision-making core is a fine-tuned vision-language model trained on multi-turn interaction traces spanning ambiguous requests, clarifications, corrections, interruptions, mixed social and task dialogue, and multi-step tasks. A purpose-built harness manages the model-tool interaction loop, while specialized navigation and manipulation policies execute physical skills. Ludi${}_{\scriptscriptstyle 0.1}$ demonstrates a practical path toward fluid human-robot collaboration today while producing the multimodal interaction traces needed to develop a more deeply integrated foundation model for robots and people.

cs.RO↗

arXiv2026

Understanding Why Language Models Hallucinate: Testing Reasoning Against Priors

Large language models often produce hallucinated answers that violate prompt-level constraints. A key diagnostic question is whether these failures reflect missing knowledge, or whether the model has the relevant information but follows the wrong inference path. We study this phenomenon as inference misalignment: a mismatch between the answer supported by the prompt and the answer favored by statistically salient latent associations. We formalize this view with a latent key-task model, in which pretraining-frequency imbalance can cause a shortcut path to dominate the constraint-sensitive path and induce positive inference loss. The framework predicts two failure modes: task-retrieval bias in entity disambiguation and key-selection bias in action choice. We introduce TrapQA, a controlled diagnostic testbed with two components. ScientistQA tests disambiguation among similar scientists with supplementary factual probes, while Real-Life Constrained QA tests everyday constraint following under salient shortcuts. Our results show that hallucination can arise from biased latent inference rather than absent knowledge alone.

cs.CL↗

arXiv2025

Semi-Automated Data Annotation in Multisensor Datasets for Autonomous Vehicle Testing

This report presents the design and implementation of a semi-automated data annotation pipeline developed within the DARTS project, whose goal is to create a large-scale, multimodal dataset of driving scenarios recorded in Polish conditions. Manual annotation of such heterogeneous data is both costly and time-consuming. To address this challenge, the proposed solution adopts a human-in-the-loop approach that combines artificial intelligence with human expertise to reduce annotation cost and duration. The system automatically generates initial annotations, enables iterative model retraining, and incorporates data anonymization and domain adaptation techniques. At its core, the tool relies on 3D object detection algorithms to produce preliminary annotations. Overall, the developed tools and methodology result in substantial time savings while ensuring consistent, high-quality annotations across different sensor modalities. The solution directly supports the DARTS project by accelerating the preparation of large annotated dataset in the project's standardized format, strengthening the technological base for autonomous vehicle research in Poland.

cs.AI↗

arXiv2025

Semi-Supervised Diversity-Aware Domain Adaptation for 3D Object detection

3D object detectors are fundamental components of perception systems in autonomous vehicles. While these detectors achieve remarkable performance on standard autonomous driving benchmarks, they often struggle to generalize across different domains - for instance, a model trained in the U.S. may perform poorly in regions like Asia or Europe. This paper presents a novel lidar domain adaptation method based on neuron activation patterns, demonstrating that state-of-the-art performance can be achieved by annotating only a small, representative, and diverse subset of samples from the target domain if they are correctly selected. The proposed approach requires very small annotation budget and, when combined with post-training techniques inspired by continual learning prevent weight drift from the original model. Empirical evaluation shows that the proposed domain adaptation approach outperforms both linear probing and state-of-the-art domain adaptation techniques.

cs.CV↗

arXiv2025

How and Why LLMs Generalize: A Fine-Grained Analysis of LLM Reasoning from Cognitive Behaviors to Low-Level Patterns

Large Language Models (LLMs) display strikingly different generalization behaviors: supervised fine-tuning (SFT) often narrows capability, whereas reinforcement-learning (RL) tuning tends to preserve it. The reasons behind this divergence remain unclear, as prior studies have largely relied on coarse accuracy metrics. We address this gap by introducing a novel benchmark that decomposes reasoning into atomic core skills such as calculation, fact retrieval, simulation, enumeration, and diagnostic, providing a concrete framework for addressing the fundamental question of what constitutes reasoning in LLMs. By isolating and measuring these core skills, the benchmark offers a more granular view of how specific cognitive abilities emerge, transfer, and sometimes collapse during post-training. Combined with analyses of low-level statistical patterns such as distributional divergence and parameter statistics, it enables a fine-grained study of how generalization evolves under SFT and RL across mathematical, scientific reasoning, and non-reasoning tasks. Our meta-probing framework tracks model behavior at different training stages and reveals that RL-tuned models maintain more stable behavioral profiles and resist collapse in reasoning skills, whereas SFT models exhibit sharper drift and overfit to surface patterns. This work provides new insights into the nature of reasoning in LLMs and points toward principles for designing training strategies that foster broad, robust generalization.

cs.LG↗

arXiv2025

IMProofBench: Benchmarking AI on Research-Level Mathematical Proof Generation

As the mathematical capabilities of large language models (LLMs) improve, it becomes increasingly important to evaluate their performance on research-level tasks at the frontier of mathematical knowledge. However, existing benchmarks are limited, as they focus solely on final-answer questions or high-school competition problems. To address this gap, we introduce IMProofBench, a private benchmark consisting of 77 peer-reviewed problems developed by expert mathematicians. Each problem requires a detailed proof and is paired with subproblems that have final answers, supporting both an evaluation by human experts and a large-scale quantitative analysis through automated grading. Furthermore, unlike prior benchmarks, the evaluation setup simulates a realistic research environment: models operate in an agentic framework with tools like web search for literature review and mathematical software such as SageMath. Our results show that current LLMs can already solve a significant percentage of research-level questions. IMProofBench will continue to evolve as a dynamic benchmark in collaboration with the mathematical community, ensuring its relevance for evaluating the next generation of LLMs.

cs.CL↗

arXiv2025

Estimating the Self-Consistency of LLMs

Systems often repeat the same prompt to large language models (LLMs) and aggregate responses to improve reliability. This short note analyzes an estimator of the self-consistency of LLMs and the tradeoffs it induces under a fixed compute budget $B=mn$, where $m$ is the number of prompts sampled from the task distribution and $n$ is the number of repeated LLM calls per prompt; the resulting analysis favors a rough split $m,n\propto\sqrt{B}$.

cs.AI↗

arXiv2025

Metric Learning in an RKHS

Metric learning from a set of triplet comparisons in the form of "Do you think item h is more similar to item i or item j?", indicating similarity and differences between items, plays a key role in various applications including image retrieval, recommendation systems, and cognitive psychology. The goal is to learn a metric in the RKHS that reflects the comparisons. Nonlinear metric learning using kernel methods and neural networks have shown great empirical promise. While previous works have addressed certain aspects of this problem, there is little or no theoretical understanding of such methods. The exception is the special (linear) case in which the RKHS is the standard Euclidean space $\mathbb{R}^d$; there is a comprehensive theory for metric learning in $\mathbb{R}^d$. This paper develops a general RKHS framework for metric learning and provides novel generalization guarantees and sample complexity bounds. We validate our findings through a set of simulations and experiments on real datasets. Our code is publicly available at https://github.com/RamyaLab/metric-learning-RKHS.

stat.ML↗

arXiv2025

Which LLMs Get the Joke? Probing Non-STEM Reasoning Abilities with HumorBench

We present HumorBench, a benchmark designed to evaluate large language models' (LLMs) ability to reason about and explain sophisticated humor in cartoon captions. As reasoning models increasingly saturate existing benchmarks in mathematics and science, novel and challenging evaluations of model intelligence beyond STEM domains are essential. Reasoning is fundamentally involved in text-based humor comprehension, requiring the identification of connections between concepts in cartoons/captions and external cultural references, wordplays, and other mechanisms. HumorBench includes approximately 300 unique cartoon-caption pairs from the New Yorker Caption Contest and Cartoonstock.com, with expert-annotated evaluation rubrics identifying essential joke elements. LLMs are evaluated based on their explanations towards the humor and abilities in identifying the joke elements. To perform well on this task, models must form and test hypotheses about associations between concepts, potentially backtracking from initial interpretations to arrive at the most plausible explanation. Our extensive benchmarking of current SOTA models reveals three key insights: (1) LLM progress on STEM reasoning transfers effectively to humor comprehension; (2) models trained exclusively on STEM reasoning data still perform well on HumorBench, demonstrating strong transferability of reasoning abilities; and (3) test-time scaling by increasing thinking token budgets yields mixed results across different models in humor reasoning.

cs.CL↗

arXiv2025

Quality evaluation of Tabby coding assistant using real source code snippets

Large language models have become a popular tool in software development, providing coding assistance. The proper measurement of the accuracy and reliability of the code produced by such tools is a challenge due to natural language prompts. We propose a simple pipeline that uses state-of-the-art implementation of classic and universal genres of algorithms and data structures. We focus on measuring the quality of TabbyML code assistant due to its open licence and the flexibility in the choice of the language model. Our results presented as cyclomatic complexity, Halstead's Bugs \& Effort and four text-based similarity matrices depict the usability of TabbyML in coding assistance tasks.

cs.SE↗

arXiv2025

BowelRCNN: Region-based Convolutional Neural Network System for Bowel Sound Auscultation

Sound events representing intestinal activity detection is a diagnostic tool with potential to identify gastrointestinal conditions. This article introduces BowelRCNN, a novel bowel sound detection system that uses audio recording, spectrogram analysys and region-based convolutional neural network (RCNN) architecture. The system was trained and validated on a real recording dataset gathered from 19 patients, comprising 60 minutes of prepared and annotated audio data. BowelRCNN achieved a classification accuracy of 96% and an F1 score of 71%. This research highlights the feasibility of using CNN architectures for bowel sound auscultation, achieving results comparable to those of recurrent-convolutional methods.

cs.SD↗

arXiv2025

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit

In-context learning is a remarkable capability of transformers, referring to their ability to adapt to specific tasks based on a short history or context. Previous research has found that task-specific information is locally encoded within models, though their emergence and functionality remain unclear due to opaque pre-training processes. In this work, we investigate the formation of task vectors in a controlled setting, using models trained from scratch on synthetic datasets. Our findings confirm that task vectors naturally emerge under certain conditions, but the tasks may be relatively weakly and/or non-locally encoded within the model. To promote strong task vectors encoded at a prescribed location within the model, we propose an auxiliary training mechanism based on a task vector prompting loss (TVP-loss). This method eliminates the need to search for task-correlated encodings within the trained model and demonstrably improves robustness and generalization.

cs.LG↗

arXiv2024

Deep Active Learning in the Open World

Machine learning models deployed in open-world scenarios often encounter unfamiliar conditions and perform poorly in unanticipated situations. As AI systems advance and find application in safety-critical domains, effectively handling out-of-distribution (OOD) data is crucial to building open-world learning systems. In this work, we introduce ALOE, a novel active learning algorithm for open-world environments designed to enhance model adaptation by incorporating new OOD classes via a two-stage approach. First, diversity sampling selects a representative set of examples, followed by energy-based OOD detection to prioritize likely unknown classes for annotation. This strategy accelerates class discovery and learning, even under constrained annotation budgets. Evaluations on three long-tailed image classification benchmarks demonstrate that ALOE outperforms traditional active learning baselines, effectively expanding known categories while balancing annotation cost. Our findings reveal a crucial tradeoff between enhancing known-class performance and discovering new classes, setting the stage for future advancements in open-world machine learning.

cs.LG↗

arXiv2024

AHA: Human-Assisted Out-of-Distribution Generalization and Detection

Modern machine learning models deployed often encounter distribution shifts in real-world applications, manifesting as covariate or semantic out-of-distribution (OOD) shifts. These shifts give rise to challenges in OOD generalization and OOD detection. This paper introduces a novel, integrated approach AHA (Adaptive Human-Assisted OOD learning) to simultaneously address both OOD generalization and detection through a human-assisted framework by labeling data in the wild. Our approach strategically labels examples within a novel maximum disambiguation region, where the number of semantic and covariate OOD data roughly equalizes. By labeling within this region, we can maximally disambiguate the two types of OOD data, thereby maximizing the utility of the fixed labeling budget. Our algorithm first utilizes a noisy binary search algorithm that identifies the maximal disambiguation region with high probability. The algorithm then continues with annotating inside the identified labeling region, reaping the full benefit of human feedback. Extensive experiments validate the efficacy of our framework. We observed that with only a few hundred human annotations, our method significantly outperforms existing state-of-the-art methods that do not involve human assistance, in both OOD generalization and OOD detection. Code is publicly available at \url{https://github.com/HaoyueBaiZJU/aha}.

cs.LG↗

arXiv2024

GPT-4o as the Gold Standard: A Scalable and General Purpose Approach to Filter Language Model Pretraining Data

Large language models require vast amounts of high-quality training data, but effective filtering of web-scale datasets remains a significant challenge. This paper demonstrates that GPT-4o is remarkably effective at identifying high-quality training data, but its prohibitive cost makes it impractical at web-scale. We propose SIEVE, a lightweight alternative that matches GPT-4o accuracy at less than 1\% of the cost. SIEVE can perform up to 500 filtering operations for the cost of one GPT-4o filtering call. The key to SIEVE is a seamless integration of GPT-4o and lightweight text classification models, using active learning to fine-tune these models in the background with a small number of calls to GPT-4o. Once trained, it performs as well as GPT-4o at a tiny fraction of the cost. Through different filtering prompts, SIEVE can efficiently curate high quality data for general or specialized domains from web-scale corpora -- a valuable capability given the current scarcity of high-quality domain-specific datasets. Extensive experiments using automatic and human evaluation metrics show that SIEVE and GPT-4o achieve similar performance on five highly specific filtering prompts. In addition, when performing quality filtering on web crawl datasets, we demonstrate SIEVE can further improve over state-of-the-art quality filtering methods in the DataComp-LM challenge for selecting LLM pretraining data.

cs.CL↗

arXiv2024

Humor in AI: Massive Scale Crowd-Sourced Preferences and Benchmarks for Cartoon Captioning

We present a novel multimodal preference dataset for creative tasks, consisting of over 250 million human ratings on more than 2.2 million captions, collected through crowdsourcing rating data for The New Yorker's weekly cartoon caption contest over the past eight years. This unique dataset supports the development and evaluation of multimodal large language models and preference-based fine-tuning algorithms for humorous caption generation. We propose novel benchmarks for judging the quality of model-generated captions, utilizing both GPT4 and human judgments to establish ranking-based evaluation strategies. Our experimental results highlight the limitations of current fine-tuning methods, such as RLHF and DPO, when applied to creative tasks. Furthermore, we demonstrate that even state-of-the-art models like GPT4 and Claude currently underperform top human contestants in generating humorous captions. As we conclude this extensive data collection effort, we release the entire preference dataset to the research community, fostering further advancements in AI humor generation and evaluation.

cs.LG↗

arXiv2024

Pretraining Decision Transformers with Reward Prediction for In-Context Multi-task Structured Bandit Learning

We study learning to learn for the multi-task structured bandit problem where the goal is to learn a near-optimal algorithm that minimizes cumulative regret. The tasks share a common structure and an algorithm should exploit the shared structure to minimize the cumulative regret for an unseen but related test task. We use a transformer as a decision-making algorithm to learn this shared structure from data collected by a demonstrator on a set of training task instances. Our objective is to devise a training procedure such that the transformer will learn to outperform the demonstrator's learning algorithm on unseen test task instances. Prior work on pretraining decision transformers either requires privileged information like access to optimal arms or cannot outperform the demonstrator. Going beyond these approaches, we introduce a pre-training approach that trains a transformer network to learn a near-optimal policy in-context. This approach leverages the shared structure across tasks, does not require access to optimal actions, and can outperform the demonstrator. We validate these claims over a wide variety of structured bandit problems to show that our proposed solution is general and can quickly identify expected rewards on unseen test tasks to support effective exploration.

cs.LG↗

arXiv2024

SaVeR: Optimal Data Collection Strategy for Safe Policy Evaluation in Tabular MDP

In this paper, we study safe data collection for the purpose of policy evaluation in tabular Markov decision processes (MDPs). In policy evaluation, we are given a \textit{target} policy and asked to estimate the expected cumulative reward it will obtain. Policy evaluation requires data and we are interested in the question of what \textit{behavior} policy should collect the data for the most accurate evaluation of the target policy. While prior work has considered behavior policy selection, in this paper, we additionally consider a safety constraint on the behavior policy. Namely, we assume there exists a known default policy that incurs a particular expected cost when run and we enforce that the cumulative cost of all behavior policies ran is better than a constant factor of the cost that would be incurred had we always run the default policy. We first show that there exists a class of intractable MDPs where no safe oracle algorithm with knowledge about problem parameters can efficiently collect data and satisfy the safety constraints. We then define the tractability condition for an MDP such that a safe oracle algorithm can efficiently collect data and using that we prove the first lower bound for this setting. We then introduce an algorithm SaVeR for this problem that approximates the safe oracle algorithm and bound the finite-sample mean squared error of the algorithm while ensuring it satisfies the safety constraint. Finally, we show in simulations that SaVeR produces low MSE policy evaluation while satisfying the safety constraint.

cs.LG↗