In this paper, we present our 3rd place system in the AVerImaTeC shared task, which combines our last year's retrieval-augmented generation (RAG) pipeline with a reverse image search (RIS) module. Despite its simplicity, our system delivers competitive performance with a single multimodal LLM call per fact-check at just $0.013 on average using GPT5.1 via OpenAI Batch API. Our system is also easy to reproduce and tweak, consisting of only three decoupled modules - a textual retrieval module based on similarity search, an image retrieval module based on API-accessed RIS, and a generation module using GPT5.1 - which is why we suggest it as an accesible starting point for further experimentation. We publish its code and prompts, as well as our vector stores and insights into the scheme's running costs and directions for further improvement.

In this paper, we present our fact verification pipeline which has scored first in FEVER 8 shared task in real-world automated factchecking. Our system is a simple two-step RAG pipeline based on our last year's submission. We show how the pipeline can be redeployed on-premise, achieving state-of-the-art fact-checking performance (in sense of Ev2R test-score), even under the constraint of a single Nvidia A10 GPU, 23GB of graphical memory and 60s running time per claim.

In this paper, we explore the problem of Claim Extraction using one-to-many text generation methods, comparing LLMs, small summarization models finetuned for the task, and a previous NER-centric baseline QACG. As the current publications on Claim Extraction, Fact Extraction, Claim Generation and Check-worthy Claim Detection are quite scattered in their means and terminology, we compile their common objectives, releasing the FEVERFact dataset, with 17K atomic factual claims extracted from 4K contextualised Wikipedia sentences, adapted from the original FEVER. We compile the known objectives into an Evaluation framework of: Atomicity, Fluency, Decontextualization, Faithfulness checked for each generated claim separately, and Focus and Coverage measured against the full set of predicted claims for a single input. For each metric, we implement a scale using a reduction to an already-explored NLP task. We validate our metrics against human grading of generic claims, to see that the model ranking on $F_{fact}$, our hardest metric, did not change and the evaluation framework approximates human grading very closely in terms of $F_1$ and RMSE.

This paper describes our $3^{rd}$ place submission in the AVeriTeC shared task in which we attempted to address the challenge of fact-checking with evidence retrieved in the wild using a simple scheme of Retrieval-Augmented Generation (RAG) designed for the task, leveraging the predictive power of Large Language Models. We release our codebase and explain its two modules - the Retriever and the Evidence & Label generator - in detail, justifying their features such as MMR-reranking and Likert-scale confidence estimation. We evaluate our solution on AVeriTeC dev and test set and interpret the results, picking the GPT-4o as the most appropriate model for our pipeline at the time of our publication, with Llama 3.1 70B being a promising open-source alternative. We perform an empirical error analysis to see that faults in our predictions often coincide with noise in the data or ambiguous fact-checks, provoking further research and data augmentation.

This article presents a pipeline for automated fact-checking leveraging publicly available Language Models and data. The objective is to assess the accuracy of textual claims using evidence from a ground-truth evidence corpus. The pipeline consists of two main modules -- the evidence retrieval and the claim veracity evaluation. Our primary focus is on the ease of deployment in various languages that remain unexplored in the field of automated fact-checking. Unlike most similar pipelines, which work with evidence sentences, our pipeline processes data on a paragraph level, simplifying the overall architecture and data requirements. Given the high cost of annotating language-specific fact-checking training data, our solution builds on the Question Answering for Claim Generation (QACG) method, which we adapt and use to generate the data for all models of the pipeline. Our strategy enables the introduction of new languages through machine translation of only two fixed datasets of moderate size. Subsequently, any number of training samples can be generated based on an evidence corpus in the target language. We provide open access to all data and fine-tuned models for Czech, English, Polish, and Slovak pipelines, as well as to our codebase that may be used to reproduce the results.We comprehensively evaluate the pipelines for all four languages, including human annotations and per-sample difficulty assessment using Pointwise V-information. The presented experiments are based on full Wikipedia snapshots to promote reproducibility. To facilitate implementation and user interaction, we develop the FactSearch application featuring the proposed pipeline and the preliminary feedback on its performance.

This interdisciplinary chapter, based on the collaboration of international lawyers and artificial intelligence (AI) experts, introduces various phases of the AI life cycle in light of possible human rights violations that may arise from each of them. It identifies the root cause of the risks to human rights and analyses the possible remedies that are common to all AI systems despite their great diversity and domain of use today. The risk of human rights violation arises notably due to unbalanced or biased data, insufficiently identified system boundary conditions or modified context, the existence of the black box, malicious use, or abuse of AI. The gist is to introduce the human rights risk assessment throughout the whole AI life cycle and integrate it into the user requirements and the system specifications in the initial phase. This ensures, inter alia, that the AI system will be developed, tested, and monitored in light of the applicable human rights limitations. Requirements related to transparency, explainability, certification, or the selection of development data are all highly relevant for protection of human rights.

In this paper, we examine several methods of acquiring Czech data for automated fact-checking, which is a task commonly modeled as a classification of textual claim veracity w.r.t. a corpus of trusted ground truths. We attempt to collect sets of data in form of a factual claim, evidence within the ground truth corpus, and its veracity label (supported, refuted or not enough info). As a first attempt, we generate a Czech version of the large-scale FEVER dataset built on top of Wikipedia corpus. We take a hybrid approach of machine translation and document alignment; the approach and the tools we provide can be easily applied to other languages. We discuss its weaknesses, propose a future strategy for their mitigation and publish the 127k resulting translations, as well as a version of such dataset reliably applicable for the Natural Language Inference task—the CsFEVER-NLI. Furthermore, we collect a novel dataset of 3,097 claims, which is annotated using the corpus of 2.2 M articles of Czech News Agency. We present an extended dataset annotation methodology based on the FEVER approach, and, as the underlying corpus is proprietary, we also publish a standalone version of the dataset for the task of Natural Language Inference we call CTKFactsNLI. We analyze both acquired datasets for spurious cues—annotation patterns leading to model overfitting. CTKFacts is further examined for inter-annotator agreement, thoroughly cleaned, and a typology of common annotator errors is extracted. Finally, we provide baseline models for all stages of the fact-checking pipeline and publish the NLI datasets, as well as our annotation platform and other experimental data.

In this paper we present two Czech datasets aimed for training automated fact-checking machine learning models. Specifically we deal with the task of assessment of a textual claim veracity w.r.t. to a (presumably) verified corpus. The output of the system is the claim classification SUPPORTS or REFUTES complemented with evidence documents or NEI (Not Enough Info) alone. In the first place we publish CsFEVER of approximately 112k claims which is an automatically generated Czech version of the well-known Wikipedia-based FEVER dataset. We took a hybrid approach of machine translation and language alignment, where the same method (and tools we provide) can be easily applied to other languages. The second dataset CTKFacts of 3,097 claims is built on the corpus of approximately two million Czech News Agency news reports. We present an extended methodology based on the FEVER approach. Most notably, we describe a method to automatically generate wider claim contexts (dictionaries) for non-hyperlinked corpora. The datasets are analyzed 1 ar X iv :2 20 1. 11 11 5v 1 [ cs .C L ] 2 6 Ja n 20 22 Springer Nature 2021 LTEX template 2 CsFEVER and CTKFacts: Czech datasets for fact verification for spurious cues, which are annotation patterns leading to model overfitting. CTKFacts is further examined for inter-annotator agreement, and a typology of common annotator errors is extracted. Finally, we provide baseline models for all stages of the fact-checking pipeline.

Dubins tours represent a solution of the Dubins traveling salesman problem (DTSP) that is a variant of the optimization routing problem to determine a curvature-constrained shortest path to visit a set of locations such that the path is feasible for Dubins vehicle, which moves only forward and has a limited turning radius. The DTSP combines the NP-hard combinatorial optimization to determine the optimal sequence of visits to the locations, as in the regular TSP, with the continuous optimization of the heading angles at the locations, where the optimal heading values depend on the sequence of visits and vice versa. We address the computationally challenging DTSP by fast evaluation of the sequence of visits by the proposed windowing surrogate model (WiSM), which estimates the length of the optimal Dubins path connecting a sequence of locations in a Dubins tour. The estimation is sped up by a regression model trained using close to optimum solutions of small Dubins tours that are generalized for large-scale instances of the addressed DTSP utilizing the sliding-window technique and a cache for already computed results. The reported results support that the proposed WiSM enables fast convergence of a relatively simple evolutionary algorithm to high-quality solutions of the DTSP. We show that with an increasing number of locations, our algorithm scales significantly better than other state-of-the-art DTSP solvers.

Two challenges can be found in a life-long classifier that learns continually: the concept drift, when the probability distribution of data is changing in time, and catastrophic forgetting when the earlier learned knowledge is lost. There are many proposed solutions to each challenge, but very little research is done to solve both challenges simultaneously. We show that both the concept drift and catastrophic forgetting are closely related to our proposed description of the life-long continual classification. We describe the process of continual learning as a wrap modification, where a wrap is a manifold that can be trained to cover or uncover a given set of samples. The notion of wraps and their cover/uncover modifiers are theoretical building blocks of a novel general life-long learning scheme, implemented as an ensemble of variational autoencoders. The proposed algorithm is examined on evaluation scenarios for continual learning and compared to state-of-the-art algorithms demonstrating the robustness to catastrophic forgetting and adaptability to concept drift but also showing the new challenges of the life-long classification.

This thesis is dedicated to methods of anomaly detection applied to log files. The current state-of-the-art anomaly detection methods usually follow the traditional approach for log processing. Firstly, log files are processed by a log parsing technique which transforms text information into non-specific structured data. Next, the data is converted into a numerical representation. The feature extraction is often related to natural language processing techniques. However, the traditional approach requires extensive domain knowledge and retraining a particular model when new log messages become available. Thanks to the recent advancements in the natural language processing domain, we can directly learn embedding vectors instead of the feature extraction based on log parsing. We propose novel autoencoder-based models leveraging the embedding vectors since autoencoders are a recommended choice in the field of anomaly detection. Moreover, we experiment with various techniques which are incorporated into autoencoders, such as convolutional layers and the self-attention mechanism. We verify that the autoencoders utilizing convolutional layers are effective for anomaly detection in log files. Furthermore, we demonstrate that boosting the models with the self-attention mechanism might be advantageous and open room for future work and further research. Finally, we can conclude that the traditional approach combined with an autoencoder may achieve impressive results on the provided testing data set. Nonetheless, the AECNN1D model achieves the most promising results among models leveraging the embedding representation of logs — the F1-score is 0.8597 on the testing data set. The AECNN1D model is generally applicable to deploying into the production since no additional requirements or periodic retraining is necessary.

Multi-goal curvature-constrained planning such as the Dubins Traveling Salesman Problem (DTSP) combines NP-hard combinatorial routing with continuous optimization to determine the optimal vehicle heading angle for each target location. The problem can be addressed as combinatorial routing using a finite set of heading samples at target locations. In such a case, optimal heading samples can be determined for a sequence of targets in polynomial time, and the DTSP can be solved as searching for a sequence with the minimal cost. However, the examination of sequences can be computationally demanding for high numbers of heading samples and target locations. A fast rejection schema is proposed to quickly examine unfavorable sequences using lower bound estimation of Dubins tour cost based on windowing technique that evaluates short subtours of the sequences. Furthermore, the computation using small problem instances can benefit from reusing stored results and thus speed up the search. The reported results indicate that the computational burden is decreased about two orders of magnitude, and the proposed approach supports finding high-quality solutions of routing problems with Dubins vehicle.

Evaluation of incremental classification algorithms is a complex task because there are many aspects to evaluate. Besides the aspects such as accuracy and generalization that are usually evaluated in the context of classification, we also need to assess how the algorithm handles two main challenges of the incremental learning: the concept drift and the catastrophic forgetting. However, only catastrophic forgetting is evaluated by the current methodology, where the classifier is evaluated in two scenarios for class addition and expansion. We generalize the methodology by proposing two new scenarios of incremental learning for class inclusion and separation that evaluate the handling of the concept drift. We demonstrate the proposed methodology on the evaluation of three different incremental classifiers, where we show that the proposed methodology provides a more complete and finer evaluation.

Activity-based modelling is a modern agent-based approach to travel demand modelling, in which the transport demand is derived from the agent's needs to perform certain activities at specific places and times. The agent's mobility is considered in a broader context, which allows the activity-based models to produce more realistic trip chains, compared to traditional trip based models. The core of any activity-based model is an activity scheduler - a software component producing sequences of agent's daily activities interconnected by trips, called activity schedules. Traditionally, activity schedulers used to rely heavily on hard-coded knowledge of transport behaviour experts. We introduce the concept of a Data-Driven Activity Scheduler (DDAS), which replaces numerous expert-designed components and their intricately engineered interactions with a collection of machine learning models. Its architecture is significantly simpler, making it easier to deploy and maintain. This shift towards data-driven, machine learning based approach is possible due to increased availability of mobility-related data. We demonstrate DDAS concept using our own proof-of-concept implementation, perform a rigorous analysis and compare the validity of the resulting model to one of the rule-based alternatives using the Validation Framework for Activity-Based Models (VALFRAM).

A life-long classifier that learns incrementally has many challenges such as concept drift, when the class changes in time, and catastrophic forgetting when the earlier learned knowledge is lost. Many successful connectionist solutions are based on an idea that new data are learned only in a part of a network that is relevant to the new data. We leverage this idea and propose a novel method for learning an ensemble of specialized autoencoders. We interpret autoencoders as manifolds that can be trained to contain or exclude given points from the input space. This manifold manipulation allows us to implement a classifier that can suppress catastrophic forgetting and adapt to concept drift. The proposed algorithm is evaluated on an incremental version of the XOR problem and on an incremental version of the MNIST classification where we achieved 0.9 accuracy which is a significant improvement over the previously published results.

On-demand transport has been disrupted by Uber and other providers, which are challenging the traditional approach adopted by taxi services. Instead of using fixed passenger pricing and driver payments, there is now the possibility of adaptation to changes in demand and supply. Properly designed, this new approach can lead to desirable tradeoffs between passenger prices, individual driver profits and provider revenue. However, pricing and allocations—known as mechanisms—are challenging problems falling in the intersection of economics and computer science. In this paper, we develop a general framework to classify mechanisms in ondemand transport. Moreover, we show that data is key to optimizing each mechanism and analyze a dataset provided by a real-world on-demand transport provider. This analysis provides valuable new insights into efficient pricing and allocation in on-demand transport. Received on 09 January 2018; accepted on 30 March 2018; published on 27 June 2018

On-demand transport has been disrupted by Uber and other providers, which are challenging the traditional approach adopted by taxi services. Instead of using fixed passenger pricing and driver payments, there is now the possibility of adaptation to changes in demand and supply. Properly designed, this new approach can lead to desirable tradeos between passenger prices, individual driver profits and provider revenue. However, pricing and allocations—known as mechanisms—are challenging problems falling in the intersection of economics and computer science. In this paper, we develop a general framework to classify mechanisms in on-demand transport. Moreover, we show that data is key to optimizing each mechanism and analyze a dataset provided by a real-world on-demand transport provider. This analysis provides valuable new insights into eÿcient pricing and allocation in on-demand transport.

Terrain classification is a crucial feature for mobile robots operating across multiple terrains. One way to learn a terrain classifier is to use a stream of labeled proprioceptive data recorded during a terrain traversal. In this paper, we propose a new terrain classifier that combines a feature extraction from a data stream with the long short-term memory (LSTM) network. Features are extracted from the information-sparse data stream by applying a sliding window computing three central moments. The feature sequence is continuously classified by the LSTM network into multiple terrain classes. Furthermore, a modified bagging method is used to deal with a limited and unbalanced training set. In comparison to the previous work on terrain classifiers for a hexapod crawling robot using only servo-drive feedback, the proposed classifier provides continuous classification with the F1 score up to 0.88, and thus provide better results than SVM classifier learned on the same input data.

Activity-based models are a specific type of agent-based models widely used in transport and urban planning to generate and study travel demand. They deal with agents that structure their behaviour in terms of daily activity schedules: sequences of activity instances (such as work, sleep or shopping) with assigned start times, durations and locations, and interconnected by trips with assigned transport modes and routes. Despite growing importance of activity-based models in transport modelling, there has been no work focus- ing specifically on statistical validation of such models so far. In this paper, we propose a six-step Validation FrameworkforActivity-basedModels(VALFRAM)thatexploitshistoricalreal-worlddatatoquantifythemodel's validityintermsofasetofnumericmetrics. Theframeworkcomparesthetemporalandspatialpropertiesand the structure of modelled activity schedules against real-world origin-destination matrices and travel diaries. We demonstrate the usefulness of the framework on a set of six dierent activity-based transport models.