Sherzod Hakimov

Sherzod Hakimov, Mattia D'Agostini, Ivan Samodelkin et al.

We introduce the Image Reconstruction Game, a fully automated benchmark in which a vision-language model issues corrective instructions to an image generator across multiple turns, making accumulated common ground directly observable as a rendered image. Benchmarking two Describer models crossed with two Generator models across seven image categories, we find that the describer is the dominant factor in reconstruction quality, while the generator determines whether iterative refinement helps or hurts. Mathematical and geometric images pose the greatest challenge. The describer's token budget strongly affects convergence: shorter budgets yield sparser first renderings with more room for visible improvement, while longer budgets raise absolute quality but leave less to fix. Stronger describers use a richer correction vocabulary spanning spatial, numeric, and structural categories, while weaker describers concentrate on surface properties and tend to stop after a few turns. Human validation shows that the best automated judge reaches only slight-to-fair agreement with human preferences, and automated scores require human recalibration to be used reliably.

Chalamalasetti Kranti, Sherzod Hakimov, David Schlangen

Robots operating in diverse environments rely on visual input to interpret objects and spatial layouts. In human-collaborative tasks, they are expected to communicate this understanding through language. Vision-language models (VLMs) support robotic tasks involving visual interpretation, question answering, and instruction following, but their capabilities in collaborative dialogue tasks requiring spatial reasoning remain underexplored. We study this gap through a collaborative structure-building task that combines visual interpretation, grounding, language-guided interaction, and action generation. We develop a framework in which VLMs use dialogue to reconstruct a target structure from visual and textual inputs. We evaluate open-weight and closed VLMs across interaction settings, input modalities, and image representations. Results show that spatial reasoning over visual representations remains difficult for the evaluated VLMs. Detailed text representations of the target yield higher reconstruction success across modality conditions, while decomposed image representations improve performance. These findings reveal limits in visual spatial grounding and grounded instruction generation for collaborative VLM agents.

Gullal S. Cheema, Sherzod Hakimov, Abdul Sittar et al.

In recent years, the problem of misinformation on the web has become widespread across languages, countries, and various social media platforms. Although there has been much work on automated fake news detection, the role of images and their variety are not well explored. In this paper, we investigate the roles of image and text at an earlier stage of the fake news detection pipeline, called claim detection. For this purpose, we introduce a novel dataset, MM-Claims, which consists of tweets and corresponding images over three topics: COVID-19, Climate Change and broadly Technology. The dataset contains roughly 86000 tweets, out of which 3400 are labeled manually by multiple annotators for the training and evaluation of multimodal models. We describe the dataset in detail, evaluate strong unimodal and multimodal baselines, and analyze the potential and drawbacks of current models.

Sherzod Hakimov, Gullal S. Cheema, Ralph Ewerth

The detection of offensive, hateful content on social media is a challenging problem that affects many online users on a daily basis. Hateful content is often used to target a group of people based on ethnicity, gender, religion and other factors. The hate or contempt toward women has been increasing on social platforms. Misogynous content detection is especially challenging when textual and visual modalities are combined to form a single context, e.g., an overlay text embedded on top of an image, also known as meme. In this paper, we present a multimodal architecture that combines textual and visual features in order to detect misogynous meme content. The proposed architecture is evaluated in the SemEval-2022 Task 5: MAMI - Multimedia Automatic Misogyny Identification challenge under the team name TIB-VA. Our solution obtained the best result in the Task-B where the challenge is to classify whether a given document is misogynous and further identify the main sub-classes of shaming, stereotype, objectification, and violence.

Sherzod Hakimov, Gullal S. Cheema

The ongoing Russo-Ukrainian conflict has been a subject of intense media coverage worldwide. Understanding the global narrative surrounding this topic is crucial for researchers that aim to gain insights into its multifaceted dimensions. In this paper, we present a novel multimedia dataset that focuses on this topic by collecting and processing tweets posted by news or media companies on social media across the globe. We collected tweets from February 2022 to May 2023 to acquire approximately 1.5 million tweets in 60 different languages along with their images. Each entry in the dataset is accompanied by processed tags, allowing for the identification of entities, stances, textual or visual concepts, and sentiment. The availability of this multimedia dataset serves as a valuable resource for researchers aiming to investigate the global narrative surrounding the ongoing conflict from various aspects such as who are the prominent entities involved, what stances are taken, where do these stances originate from, how are the different textual and visual concepts related to the event portrayed.

Yi-Sheng Hsu, Nils Feldhus, Sherzod Hakimov

Free-text rationales justify model decisions in natural language and thus become likable and accessible among approaches to explanation across many tasks. However, their effectiveness can be hindered by misinterpretation and hallucination. As a perturbation test, we investigate how large language models (LLMs) perform rationale generation under the effects of readability level control, i.e., being prompted for an explanation targeting a specific expertise level, such as sixth grade or college. We find that explanations are adaptable to such instruction, though the observed distinction between readability levels does not fully match the defined complexity scores according to traditional readability metrics. Furthermore, the generated rationales tend to feature medium level complexity, which correlates with the measured quality using automatic metrics. Finally, our human annotators confirm a generally satisfactory impression on rationales at all readability levels, with high-school-level readability being most commonly perceived and favored.

Chalamalasetti Kranti, Sherzod Hakimov, David Schlangen

While there has been a lot of research recently on robots in household environments, at the present time, most robots in existence can be found on shop floors, and most interactions between humans and robots happen there. ``Collaborative robots'' (cobots) designed to work alongside humans on assembly lines traditionally require expert programming, limiting ability to make changes, or manual guidance, limiting expressivity of the resulting programs. To address these limitations, we explore using Large Language Models (LLMs), and in particular, their abilities of doing in-context learning, for conversational code generation. As a first step, we define RATS, the ``Repetitive Assembly Task'', a 2D building task designed to lay the foundation for simulating industry assembly scenarios. In this task, a `programmer' instructs a cobot, using natural language, on how a certain assembly is to be built; that is, the programmer induces a program, through natural language. We create a dataset that pairs target structures with various example instructions (human-authored, template-based, and model-generated) and example code. With this, we systematically evaluate the capabilities of state-of-the-art LLMs for synthesising this kind of code, given in-context examples. Evaluating in a simulated environment, we find that LLMs are capable of generating accurate `first order code' (instruction sequences), but have problems producing `higher-order code' (abstractions such as functions, or use of loops).

Sherzod Hakimov

Natural language processing for the Turkic language family, spoken by over 200 million people across Eurasia, remains fragmented, with most languages lacking unified tooling and resources. We present TurkicNLP, an open-source Python library providing a single, consistent NLP pipeline for Turkic languages across four script families: Latin, Cyrillic, Perso-Arabic, and Old Turkic Runic. The library covers tokenization, morphological analysis, part-of-speech tagging, dependency parsing, named entity recognition, bidirectional script transliteration, cross-lingual sentence embeddings, and machine translation through one language-agnostic API. A modular multi-backend architecture integrates rule-based finite-state transducers and neural models transparently, with automatic script detection and routing between script variants. Outputs follow the CoNLL-U standard for full interoperability and extension. Code and documentation are hosted at https://github.com/turkic-nlp/turkicnlp .

Jing Yang, Nils Feldhus, Salar Mohtaj et al.

Despite advances in Natural Language Generation (NLG), evaluation remains challenging. Although various new metrics and LLM-as-a-judge (LaaJ) methods are proposed, human judgment persists as the gold standard. To systematically review how NLG evaluation has evolved, we employ an automatic information extraction scheme to gather key information from NLG papers, focusing on different evaluation methods (metrics, LaaJ and human evaluation). With extracted metadata from 14,171 papers across four major conferences (ACL, EMNLP, NAACL, and INLG) over the past six years, we reveal several critical findings: (1) Task Divergence: While Dialogue Generation demonstrates a rapid shift toward LaaJ (>40% in 2025), Machine Translation remains locked into n-gram metrics, and Question Answering exhibits a substantial decline in the proportion of studies conducting human evaluation. (2) Metric Inertia: Despite the development of semantic metrics, general-purpose metrics (e.g., BLEU, ROUGE) continue to be widely used across tasks without empirical justification, often lacking the discriminative power to distinguish between specific quality criteria. (3) Human-LaaJ Divergence: Our association analysis challenges the assumption that LLMs act as mere proxies for humans; LaaJ and human evaluations prioritize very different signals, and explicit validation is scarce (<8% of papers comparing the two), with only moderate to low correlation. Based on these observations, we derive practical recommendations to improve the rigor of future NLG evaluation.

Sherzod Hakimov, David Schlangen

Large language models have demonstrated robust performance on various language tasks using zero-shot or few-shot learning paradigms. While being actively researched, multimodal models that can additionally handle images as input have yet to catch up in size and generality with language-only models. In this work, we ask whether language-only models can be utilised for tasks that require visual input -- but also, as we argue, often require a strong reasoning component. Similar to some recent related work, we make visual information accessible to the language model using separate verbalisation models. Specifically, we investigate the performance of open-source, open-access language models against GPT-3 on five vision-language tasks when given textually-encoded visual information. Our results suggest that language models are effective for solving vision-language tasks even with limited samples. This approach also enhances the interpretability of a model's output by providing a means of tracing the output back through the verbalised image content.

Gaurish Thakkar, Sherzod Hakimov, Marko Tadić

In recent years, multimodal natural language processing, aimed at learning from diverse data types, has garnered significant attention. However, there needs to be more clarity when it comes to analysing multimodal tasks in multi-lingual contexts. While prior studies on sentiment analysis of tweets have predominantly focused on the English language, this paper addresses this gap by transforming an existing textual Twitter sentiment dataset into a multimodal format through a straightforward curation process. Our work opens up new avenues for sentiment-related research within the research community. Additionally, we conduct baseline experiments utilising this augmented dataset and report the findings. Notably, our evaluations reveal that when comparing unimodal and multimodal configurations, using a sentiment-tuned large language model as a text encoder performs exceptionally well.

Philipp Sadler, Sherzod Hakimov, David Schlangen

In collaborative goal-oriented settings, the participants are not only interested in achieving a successful outcome, but do also implicitly negotiate the effort they put into the interaction (by adapting to each other). In this work, we propose a challenging interactive reference game that requires two players to coordinate on vision and language observations. The learning signal in this game is a score (given after playing) that takes into account the achieved goal and the players' assumed efforts during the interaction. We show that a standard Proximal Policy Optimization (PPO) setup achieves a high success rate when bootstrapped with heuristic partner behaviors that implement insights from the analysis of human-human interactions. And we find that a pairing of neural partners indeed reduces the measured joint effort when playing together repeatedly. However, we observe that in comparison to a reasonable heuristic pairing there is still room for improvement -- which invites further research in the direction of cost-sharing in collaborative interactions.

Nicola Horst, Davide Mazzaccara, Antonia Schmidt et al.

Interaction between learner and feedback-giver has come into focus recently for post-training of Large Language Models (LLMs), through the use of reward models that judge the appropriateness of a model's response. In this paper, we investigate whether Dialogue Games -- goal-directed and rule-governed activities driven predominantly by verbal actions -- can also serve as a source of feedback signals for learning. We introduce Playpen, an environment for off- and online learning through Dialogue Game self-play, and investigate a representative set of post-training methods: supervised fine-tuning; direct alignment (DPO); and reinforcement learning with GRPO. We experiment with post-training a small LLM (Llama-3.1-8B-Instruct), evaluating performance on unseen instances of training games as well as unseen games, and on standard benchmarks. We find that imitation learning through SFT improves performance on unseen instances, but negatively impacts other skills, while interactive learning with GRPO shows balanced improvements without loss of skills. We release the framework and the baseline training setups to foster research in the promising new direction of learning in (synthetic) interaction.

Chalamalasetti Kranti, Sherzod Hakimov, David Schlangen

The emergence of instruction-tuned large language models (LLMs) has advanced the field of dialogue systems, enabling both realistic user simulations and robust multi-turn conversational agents. However, existing research often evaluates these components in isolation-either focusing on a single user simulator or a specific system design-limiting the generalisability of insights across architectures and configurations. In this work, we propose clem todd (chat-optimized LLMs for task-oriented dialogue systems development), a flexible framework for systematically evaluating dialogue systems under consistent conditions. clem todd enables detailed benchmarking across combinations of user simulators and dialogue systems, whether existing models from literature or newly developed ones. It supports plug-and-play integration and ensures uniform datasets, evaluation metrics, and computational constraints. We showcase clem todd's flexibility by re-evaluating existing task-oriented dialogue systems within this unified setup and integrating three newly proposed dialogue systems into the same evaluation pipeline. Our results provide actionable insights into how architecture, scale, and prompting strategies affect dialogue performance, offering practical guidance for building efficient and effective conversational AI systems.

Sherzod Hakimov, Lara Pfennigschmidt, David Schlangen

This study utilizes the game Codenames as a benchmarking tool to evaluate large language models (LLMs) with respect to specific linguistic and cognitive skills. LLMs play each side of the game, where one side generates a clue word covering several target words and the other guesses those target words. We designed various experiments by controlling the choice of words (abstract vs. concrete words, ambiguous vs. monosemic) or the opponent (programmed to be faster or slower in revealing words). Recent commercial and open-weight models were compared side-by-side to find out factors affecting their performance. The evaluation reveals details about their strategies, challenging cases, and limitations of LLMs.

Philipp Sadler, Sherzod Hakimov, David Schlangen

Albrecht and Stone (2018) state that modeling of changing behaviors remains an open problem "due to the essentially unconstrained nature of what other agents may do". In this work we evaluate the adaptability of neural artificial agents towards assumed partner behaviors in a collaborative reference game. In this game success is achieved when a knowledgeable Guide can verbally lead a Follower to the selection of a specific puzzle piece among several distractors. We frame this language grounding and coordination task as a reinforcement learning problem and measure to which extent a common reinforcement training algorithm (PPO) is able to produce neural agents (the Guides) that perform well with various heuristic Follower behaviors that vary along the dimensions of confidence and autonomy. We experiment with a learning signal that in addition to the goal condition also respects an assumed communicative effort. Our results indicate that this novel ingredient leads to communicative strategies that are less verbose (staying silent in some of the steps) and that with respect to that the Guide's strategies indeed adapt to the partner's level of confidence and autonomy.

Sherzod Hakimov, Roland Bernard, Tim Leiber et al.

Negotiation is a fundamental challenge for AI agents, as it requires an ability to reason strategically, model opponents, and balance cooperation with competition. We conduct the first comprehensive study systematically evaluating the effect of (LLM-)reasoning on the negotiation abilities of both commercial and open-weight LLMs, and do this across three languages. Using a self-play setup across three diverse dialogue games, we analyse trade-offs between performance and cost, the language consistency of reasoning processes, and the nature of strategic adaptation exhibited by models. Our findings show that enabling reasoning-that is, scaling test time compute-significantly improves negotiation outcomes by enhancing collaboration and helping models overcome task complexities, but comes at a substantial computational cost: reasoning improves GPT-5's performance by 31.4 % while increasing its cost by nearly 400 %. Most critically, we uncover a significant multilingual reasoning distinction: open-weight models consistently switch to English for their internal reasoning steps, even when negotiating in German or Italian (and thus possibly impacting potential explainability gains through the disclosure of reasoning traces), while leading commercial models maintain language consistency between their reasoning and final output.

David Schlangen, Sherzod Hakimov, Jonathan Jordan et al.

There are currently two main paradigms for evaluating large language models (LLMs), reference-based evaluation and preference-based evaluation. The first, carried over from the evaluation of machine learning models in general, relies on pre-defined task instances, for which reference task executions are available. The second, best exemplified by the LM-arena, relies on (often self-selected) users bringing their own intents to a site that routes these to several models in parallel, among whose responses the user then selects their most preferred one. The former paradigm hence excels at control over what is tested, while the latter comes with higher ecological validity, testing actual use cases interactively. Recently, a third complementary paradigm has emerged that combines some of the strengths of these approaches, offering control over multi-turn, reference-free, repeatable interactions, while stressing goal-directedness: dialogue game based evaluation. While the utility of this approach has been shown by several projects, its adoption has been held back by the lack of a mature, easily re-usable implementation. In this paper, we present clembench, which has been in continuous development since 2023 and has in its latest release been optimized for ease of general use. We describe how it can be used to benchmark one's own models (using a provided set of benchmark game instances in English), as well as how easily the benchmark itself can be extended with new, tailor-made targeted tests.

Chalamalasetti Kranti, Sherzod Hakimov, David Schlangen

Instruction-tuned large language models (LLMs) have shown strong performance on a variety of tasks; however, generalizing from synthetic to human-authored instructions in grounded environments remains a challenge for them. In this work, we study generalization challenges in spatial grounding tasks where models interpret and translate instructions for building object arrangements on a $2.5$D grid. We fine-tune LLMs using only synthetic instructions and evaluate their performance on a benchmark dataset containing both synthetic and human-written instructions. Our results reveal that while models generalize well on simple tasks, their performance degrades significantly on more complex tasks. We present a detailed error analysis of the gaps in instruction generalization.

Jonathan Jordan, Sherzod Hakimov, David Schlangen

Large Language Models (LLMs) serve not only as chatbots but as key components in agent systems, where their common-sense knowledge significantly impacts performance as language-based planners for situated or embodied action. We assess LLMs' incremental learning (based on feedback from the environment), and controlled in-context learning abilities using a text-based environment. We introduce challenging yet interesting set of experiments to test i) how agents can incrementally solve tasks related to every day objects in typical rooms in a house where each of them are discovered by interacting within the environment, ii) controlled in-context learning abilities and efficiency of agents by providing short info about locations of objects and rooms to check how faster the task can be solved, and finally iii) using synthetic pseudo-English words to gauge how well LLMs are at inferring meaning of unknown words from environmental feedback. Results show that larger commercial models have a substantial gap in performance compared to open-weight but almost all models struggle with the synthetic words experiments.

Chalamalasetti Kranti, Sherzod Hakimov, David Schlangen

In the Minecraft Collaborative Building Task, two players collaborate: an Architect (A) provides instructions to a Builder (B) to assemble a specified structure using 3D blocks. In this work, we investigate the use of large language models (LLMs) to predict the sequence of actions taken by the Builder. Leveraging LLMs' in-context learning abilities, we use few-shot prompting techniques, that significantly improve performance over baseline methods. Additionally, we present a detailed analysis of the gaps in performance for future work

Nidhir Bhavsar, Jonathan Jordan, Sherzod Hakimov et al.

What makes a good Large Language Model (LLM)? That it performs well on the relevant benchmarks -- which hopefully measure, with some validity, the presence of capabilities that are also challenged in real application. But what makes the model perform well? What gives a model its abilities? We take a recently introduced type of benchmark that is meant to challenge capabilities in a goal-directed, agentive context through self-play of conversational games, and analyse how performance develops as a function of model characteristics like number of parameters, or type of training. We find that while there is a clear relationship between number of parameters and performance, there is still a wide spread of performance points within a given size bracket, which is to be accounted for by training parameters such as fine-tuning data quality and method. From a more practical angle, we also find a certain degree of unpredictability about performance across access methods, possible due to unexposed sampling parameters, and a, very welcome, performance stability against at least moderate weight quantisation during inference.

Sherzod Hakimov, Yerkezhan Abdullayeva, Kushal Koshti et al.

While the situation has improved for text-only models, it again seems to be the case currently that multimodal (text and image) models develop faster than ways to evaluate them. In this paper, we bring a recently developed evaluation paradigm from text models to multimodal models, namely evaluation through the goal-oriented game (self) play, complementing reference-based and preference-based evaluation. Specifically, we define games that challenge a model's capability to represent a situation from visual information and align such representations through dialogue. We find that the largest closed models perform rather well on the games that we define, while even the best open-weight models struggle with them. On further analysis, we find that the exceptional deep captioning capabilities of the largest models drive some of the performance. There is still room to grow for both kinds of models, ensuring the continued relevance of the benchmark.

Sahar Tahmasebi, Sherzod Hakimov, Ralph Ewerth et al.

The increasing proliferation of misinformation and its alarming impact have motivated both industry and academia to develop approaches for fake news detection. However, state-of-the-art approaches are usually trained on datasets of smaller size or with a limited set of specific topics. As a consequence, these models lack generalization capabilities and are not applicable to real-world data. In this paper, we propose three models that adopt and fine-tune state-of-the-art multimodal transformers for multimodal fake news detection. We conduct an in-depth analysis by manipulating the input data aimed to explore models performance in realistic use cases on social media. Our study across multiple models demonstrates that these systems suffer significant performance drops against manipulated data. To reduce the bias and improve model generalization, we suggest training data augmentation to conduct more meaningful experiments for fake news detection on social media. The proposed data augmentation techniques enable models to generalize better and yield improved state-of-the-art results.

Kranti Chalamalasetti, Jana Götze, Sherzod Hakimov et al.

Recent work has proposed a methodology for the systematic evaluation of "Situated Language Understanding Agents"-agents that operate in rich linguistic and non-linguistic contexts-through testing them in carefully constructed interactive settings. Other recent work has argued that Large Language Models (LLMs), if suitably set up, can be understood as (simulators of) such agents. A connection suggests itself, which this paper explores: Can LLMs be evaluated meaningfully by exposing them to constrained game-like settings that are built to challenge specific capabilities? As a proof of concept, this paper investigates five interaction settings, showing that current chat-optimised LLMs are, to an extent, capable to follow game-play instructions. Both this capability and the quality of the game play, measured by how well the objectives of the different games are met, follows the development cycle, with newer models performing better. The metrics even for the comparatively simple example games are far from being saturated, suggesting that the proposed instrument will remain to have diagnostic value. Our general framework for implementing and evaluating games with LLMs is available at https://github.com/clembench .

Philipp Sadler, Sherzod Hakimov, David Schlangen

The ability to pick up on language signals in an ongoing interaction is crucial for future machine learning models to collaborate and interact with humans naturally. In this paper, we present an initial study that evaluates intra-episodic feedback given in a collaborative setting. We use a referential language game as a controllable example of a task-oriented collaborative joint activity. A teacher utters a referring expression generated by a well-known symbolic algorithm (the "Incremental Algorithm") as an initial instruction and then monitors the follower's actions to possibly intervene with intra-episodic feedback (which does not explicitly have to be requested). We frame this task as a reinforcement learning problem with sparse rewards and learn a follower policy for a heuristic teacher. Our results show that intra-episodic feedback allows the follower to generalize on aspects of scene complexity and performs better than providing only the initial statement.

Sherzod Hakimov, Ralph Ewerth

The detection of offensive, hateful and profane language has become a critical challenge since many users in social networks are exposed to cyberbullying activities on a daily basis. In this paper, we present an analysis of combining different textual features for the detection of hateful or offensive posts on Twitter. We provide a detailed experimental evaluation to understand the impact of each building block in a neural network architecture. The proposed architecture is evaluated on the English Subtask 1A: Identifying Hate, offensive and profane content from the post datasets of HASOC-2021 dataset under the team name TIB-VA. We compared different variants of the contextual word embeddings combined with the character level embeddings and the encoding of collected hate terms.

Gullal S. Cheema, Sherzod Hakimov, Eric Müller-Budack et al.

Opinion and sentiment analysis is a vital task to characterize subjective information in social media posts. In this paper, we present a comprehensive experimental evaluation and comparison with six state-of-the-art methods, from which we have re-implemented one of them. In addition, we investigate different textual and visual feature embeddings that cover different aspects of the content, as well as the recently introduced multimodal CLIP embeddings. Experimental results are presented for two different publicly available benchmark datasets of tweets and corresponding images. In contrast to the evaluation methodology of previous work, we introduce a reproducible and fair evaluation scheme to make results comparable. Finally, we conduct an error analysis to outline the limitations of the methods and possibilities for the future work.

Hussain Kanafani, Junaid Ahmed Ghauri, Sherzod Hakimov et al.

Video summarization aims at generating a compact yet representative visual summary that conveys the essence of the original video. The advantage of unsupervised approaches is that they do not require human annotations to learn the summarization capability and generalize to a wider range of domains. Previous work relies on the same type of deep features, typically based on a model pre-trained on ImageNet data. Therefore, we propose the incorporation of multiple feature sources with chunk and stride fusion to provide more information about the visual content. For a comprehensive evaluation on the two benchmarks TVSum and SumMe, we compare our method with four state-of-the-art approaches. Two of these approaches were implemented by ourselves to reproduce the reported results. Our evaluation shows that we obtain state-of-the-art results on both datasets, while also highlighting the shortcomings of previous work with regard to the evaluation methodology. Finally, we perform error analysis on videos for the two benchmark datasets to summarize and spot the factors that lead to misclassifications.

Golsa Tahmasebzadeh, Endri Kacupaj, Eric Müller-Budack et al.

In the context of social media, geolocation inference on news or events has become a very important task. In this paper, we present the GeoWINE (Geolocation-based Wiki-Image-News-Event retrieval) demonstrator, an effective modular system for multimodal retrieval which expects only a single image as input. The GeoWINE system consists of five modules in order to retrieve related information from various sources. The first module is a state-of-the-art model for geolocation estimation of images. The second module performs a geospatial-based query for entity retrieval using the Wikidata knowledge graph. The third module exploits four different image embedding representations, which are used to retrieve most similar entities compared to the input image. The embeddings are derived from the tasks of geolocation estimation, place recognition, ImageNet-based image classification, and their combination. The last two modules perform news and event retrieval from EventRegistry and the Open Event Knowledge Graph (OEKG). GeoWINE provides an intuitive interface for end-users and is insightful for experts for reconfiguration to individual setups. The GeoWINE achieves promising results in entity label prediction for images on Google Landmarks dataset. The demonstrator is publicly available at http://cleopatra.ijs.si/geowine/.

Junaid Ahmed Ghauri, Sherzod Hakimov, Ralph Ewerth

The assignment of importance scores to particular frames or (short) segments in a video is crucial for summarization, but also a difficult task. Previous work utilizes only one source of visual features. In this paper, we suggest a novel model architecture that combines three feature sets for visual content and motion to predict importance scores. The proposed architecture utilizes an attention mechanism before fusing motion features and features representing the (static) visual content, i.e., derived from an image classification model. Comprehensive experimental evaluations are reported for two well-known datasets, SumMe and TVSum. In this context, we identify methodological issues on how previous work used these benchmark datasets, and present a fair evaluation scheme with appropriate data splits that can be used in future work. When using static and motion features with parallel attention mechanism, we improve state-of-the-art results for SumMe, while being on par with the state of the art for the other dataset.

Gullal S. Cheema, Sherzod Hakimov, Eric Müller-Budack et al.

Fake news is a severe problem in social media. In this paper, we present an empirical study on visual, textual, and multimodal models for the tasks of claim, claim check-worthiness, and conspiracy detection, all of which are related to fake news detection. Recent work suggests that images are more influential than text and often appear alongside fake text. To this end, several multimodal models have been proposed in recent years that use images along with text to detect fake news on social media sites like Twitter. However, the role of images is not well understood for claim detection, specifically using transformer-based textual and multimodal models. We investigate state-of-the-art models for images, text (Transformer-based), and multimodal information for four different datasets across two languages to understand the role of images in the task of claim and conspiracy detection.

Gullal S. Cheema, Sherzod Hakimov, Ralph Ewerth

Fake news on social media has become a hot topic of research as it negatively impacts the discourse of real news in the public. Specifically, the ongoing COVID-19 pandemic has seen a rise of inaccurate and misleading information due to the surrounding controversies and unknown details at the beginning of the pandemic. The FakeNews task at MediaEval 2020 tackles this problem by creating a challenge to automatically detect tweets containing misinformation based on text and structure from Twitter follower network. In this paper, we present a simple approach that uses BERT embeddings and a shallow neural network for classifying tweets using only text, and discuss our findings and limitations of the approach in text-based misinformation detection.

Eric Müller-Budack, Matthias Springstein, Sherzod Hakimov et al.

Event classification can add valuable information for semantic search and the increasingly important topic of fact validation in news. So far, only few approaches address image classification for newsworthy event types such as natural disasters, sports events, or elections. Previous work distinguishes only between a limited number of event types and relies on rather small datasets for training. In this paper, we present a novel ontology-driven approach for the classification of event types in images. We leverage a large number of real-world news events to pursue two objectives: First, we create an ontology based on Wikidata comprising the majority of event types. Second, we introduce a novel large-scale dataset that was acquired through Web crawling. Several baselines are proposed including an ontology-driven learning approach that aims to exploit structured information of a knowledge graph to learn relevant event relations using deep neural networks. Experimental results on existing as well as novel benchmark datasets demonstrate the superiority of the proposed ontology-driven approach.

Junaid Ahmed Ghauri, Sherzod Hakimov, Ralph Ewerth

Videos are a commonly-used type of content in learning during Web search. Many e-learning platforms provide quality content, but sometimes educational videos are long and cover many topics. Humans are good in extracting important sections from videos, but it remains a significant challenge for computers. In this paper, we address the problem of assigning importance scores to video segments, that is how much information they contain with respect to the overall topic of an educational video. We present an annotation tool and a new dataset of annotated educational videos collected from popular online learning platforms. Moreover, we propose a multimodal neural architecture that utilizes state-of-the-art audio, visual and textual features. Our experiments investigate the impact of visual and temporal information, as well as the combination of multimodal features on importance prediction.

Gullal S. Cheema, Sherzod Hakimov, Ralph Ewerth

In this digital age of news consumption, a news reader has the ability to react, express and share opinions with others in a highly interactive and fast manner. As a consequence, fake news has made its way into our daily life because of very limited capacity to verify news on the Internet by large companies as well as individuals. In this paper, we focus on solving two problems which are part of the fact-checking ecosystem that can help to automate fact-checking of claims in an ever increasing stream of content on social media. For the first problem, claim check-worthiness prediction, we explore the fusion of syntactic features and deep transformer Bidirectional Encoder Representations from Transformers (BERT) embeddings, to classify check-worthiness of a tweet, i.e. whether it includes a claim or not. We conduct a detailed feature analysis and present our best performing models for English and Arabic tweets. For the second problem, claim retrieval, we explore the pre-trained embeddings from a Siamese network transformer model (sentence-transformers) specifically trained for semantic textual similarity, and perform KD-search to retrieve verified claims with respect to a query tweet.

Golsa Tahmasebzadeh, Sherzod Hakimov, Eric Müller-Budack et al.

Content-based information retrieval is based on the information contained in documents rather than using metadata such as keywords. Most information retrieval methods are either based on text or image. In this paper, we investigate the usefulness of multimodal features for cross-lingual news search in various domains: politics, health, environment, sport, and finance. To this end, we consider five feature types for image and text and compare the performance of the retrieval system using different combinations. Experimental results show that retrieval results can be improved when considering both visual and textual information. In addition, it is observed that among textual features entity overlap outperforms word embeddings, while geolocation embeddings achieve better performance among visual features in the retrieval task.

Sherzod Hakimov, Soufian Jebbara, Philipp Cimiano

The task of answering natural language questions over knowledge bases has received wide attention in recent years. Various deep learning architectures have been proposed for this task. However, architectural design choices are typically not systematically compared nor evaluated under the same conditions. In this paper, we contribute to a better understanding of the impact of architectural design choices by evaluating four different architectures under the same conditions. We address the task of answering simple questions, consisting in predicting the subject and predicate of a triple given a question. In order to provide a fair comparison of different architectures, we evaluate them under the same strategy for inferring the subject, and compare different architectures for inferring the predicate. The architecture for inferring the subject is based on a standard LSTM model trained to recognize the span of the subject in the question and on a linking component that links the subject span to an entity in the knowledge base. The architectures for predicate inference are based on i) a standard softmax classifier ranging over all predicates as output, iii) a model that predicts a low-dimensional encoding of the property given entity representation and question, iii) a model that learns to score a pair of subject and predicate given the question as well as iv) a model based on the well-known FastText model. The comparison of architectures shows that FastText provides better results than other architectures.

Sherzod Hakimov, Soufian Jebbara, Philipp Cimiano

The task of answering natural language questions over RDF data has received wide interest in recent years, in particular in the context of the series of QALD benchmarks. The task consists of mapping a natural language question to an executable form, e.g. SPARQL, so that answers from a given KB can be extracted. So far, most systems proposed are i) monolingual and ii) rely on a set of hard-coded rules to interpret questions and map them into a SPARQL query. We present the first multilingual QALD pipeline that induces a model from training data for mapping a natural language question into logical form as probabilistic inference. In particular, our approach learns to map universal syntactic dependency representations to a language-independent logical form based on DUDES (Dependency-based Underspecified Discourse Representation Structures) that are then mapped to a SPARQL query as a deterministic second step. Our model builds on factor graphs that rely on features extracted from the dependency graph and corresponding semantic representations. We rely on approximate inference techniques, Markov Chain Monte Carlo methods in particular, as well as Sample Rank to update parameters using a ranking objective. Our focus lies on developing methods that overcome the lexical gap and present a novel combination of machine translation and word embedding approaches for this purpose. As a proof of concept for our approach, we evaluate our approach on the QALD-6 datasets for English, German & Spanish.