Steve is an AI career coaching platform that turns a resume and insights from an AI-enabled chat with a user into a personalized skill gap report and upskilling roadmap. The platform suggests a personalized course plan, supports continuous learning, and helps shape the user’s career trajectory. Steve is built around schema-constrained JSON artifacts and a configurable career-tree ontology. The system compares confirmed skills against role-specific requirements, prioritizes gaps (critical/important/beneficial), and translates the analysis into embedding-based queries over the course index. Steve has three personas (Interview Coach, Resume Evaluator, and Career Coach) that provide concise feedback tailored to the user’s goals and context. Steve also supports speech Input/Output (I/O) via Whisper-based speech-to-text and a dual-voice text-to-speech layer, enabling users to talk to Steve. The platform offers flexible adaptability across institutions, enabling them to configure deployments by substituting their own ontologies and course catalogs. Our demo uses STEM trajectories as a case study, but the pipeline is domain-agnostic by design. Users can edit inputs, check speech recognition accuracy, and observe consistent updates, illustrating a reproducible, human-in-the-loop pattern for deploying LLMs in career guidance. Steve is currently in its alpha stage and available for demonstration.

Social media provides valuable insights into societal opinions and user interactions. Social landscapes, created from these interactions, offer a comprehensive view of online conversations and social media dynamics. The development of advanced data analytics tools has made the creation of social landscapes for larger populations increasingly common and accessible for researchers. This underscores the importance of clearly defining and organizing the insights we seek from social landscapes. We introduce a methodology to analyze the social structure through social landscapes. We have identified 12 key elements that encapsulate the insights expected from social landscapes. Then, we integrate two comprehensive social landscape approaches into our methodology that effectively provide insights into the key elements previously outlined. These two approaches illuminate different facets of social media dynamics: one focuses on the content generated and the other on relationship-based interactions. First, we revisit the concept ofgalaxiesas a single time-based snapshot of large-scale online conversations. Second, we introduce a technique that utilizes the network of user interactions on social media to map social structures. We propose a novel method to construct a sociopolitical spectrum using discourse trajectory inference (pseudotime transformation), marking its first use outside bioinformatics literature. Finally, we take the introduced methodology to evaluate how each social landscape enhances our understanding of social media dynamics. These insights can help to structure the insights we aim to extract from social landscapes and provide practical tools for media analysts and strategists aiming to analyze social media dynamics effectively.

Formally verifying properties of software code has been a highly desirable task, especially with the emergence of LLM-generated code. In the same vein, they provide an interesting avenue for the exploration of formal verification and mechanistic interpretability. Since the introduction of code-specific models, despite their successes in generating code in Lean4 and Isabelle, the task of generalized theorem proving still remains far from being fully solved and will be a benchmark for reasoning capability in LLMs. In this work, we introduce a framework that generates whole proofs in a formal language to be used within systems that utilize the power of built-in tactics and off-the-shelf automated theorem provers. Our framework includes 3 components: generating natural language statements of the code to be verified, an LLM that generates formal proofs for the given statement, and a module employing heuristics for building the final proof. To train the LLM, we employ a 2-stage fine-tuning process, where we first use SFT-based training to enable the model to generate syntactically correct Isabelle code and then RL-based training that encourages the model to generate proofs verified by a theorem prover. We validate our framework using the miniF2F-test benchmark and the Isabelle proof assistant and design a use case to verify the correctness of the AWS S3 bucket access policy code. We also curate a dataset based on the FVEL\textsubscript{\textnormal{ER}} dataset for future training tasks.

The increasingly widespread usage of the Internet and social networks has changed the way people interact with each other and react to events. These interactions enable positive collective outcomes such as enhancing collaboration in science. Conversely, undesirable effects have emerged: the self-segregation of online users within “bubbles” of biased content, the spread of misinformation, and the growing diffusion of aggressive speech with radical emotional valence. Affective polarization is the extent to which two opposing groups dislike one another, and it could be measured as the degree to which the two groups are willing to discriminate one against the other. Such social mechanism could occur in online social networks as a result of a controversial event in the offline world. To counter affective polarization, influential actors often make interventions using counter narratives in online social networks. However, a quantitative measure of the effectiveness of such counter narratives is typically not provided. In this study, we propose an approach to evaluate the affective polarization in online discussions generated by an offline controversial event and a measure of the effectiveness of counter narratives made by influential actors to attenuate the rise of affective polarization. The proposed approach was applied to five cases of controversial events that occurred in European soccer leagues using data collected from Twitter. Such an approach could be generalized to any other scenario involving an offline event that sparks divergent emotional reactions in online discussions and an official social media account that intervenes with a counter narrative to impact affective polarization.

The emergence of social media has facilitated new ways of communication and social engagement. Political elites and public figures use social media to inform and mobilize users. Discussions appearing around these tweets reflect valuable information from the communities, their interests, and concerns. The majority of the studies on conversational dynamics on social media use the network between users to track the flow of information and impact. In this study, a different approach is introduced in which discussion threads as the network between tweets are considered as the main components of the analysis. Based on this approach, a novel framework is proposed to provide a high-level overview of a Twitter stream. The metric energy is introduced to quantify user engagement in discussions initiated by each tweet and to study the patterns of user-participation overtime. To mine the content of the tweets, some recently developed text analysis methods are integrated into our model to extract the topic of the tweets, the sentiment of the tweets, the stance of tweets toward a controversial topic, and the coherency between the tweet-reply pairs. Three different data sets from politics, healthcare, and sport are collected to be used as case studies. The proposed framework provides a better understanding of how opinions form around a tweet, how different topics to bond together and make the whole discussions, for which topics discussions are more coherent, and which users are able to initiate a cross cutting and coherent conversation over social media.