Large language models are often equipped with safety alignment mechanisms designed to prevent generation of harmful or other unwanted content. However, an increasing number of jailbreaking techniques attempt to circumvent these safeguards, raising significant safety concerns. This paper introduces an open-source evaluation framework that analyzes jailbreaking effectiveness in several dimensions: refusal bypass rate, harmful response quality, impact on general model capabilities, and computational cost. In the study, prompt injection, sampling exploits, and model manipulation techniques are examined across four open-weight instruction-tuned large language models. The results demonstrate that high refusal bypass does not necessarily equate to practical safety compromise. Specifically, model manipulation methods like single refusal direction ablation achieve a high attack success rate, but often degrade general capabilities and require significant computational resources. Meanwhile, sampling-based exploits show a minimal practical threat when assessed with a robust model classifier. The findings emphasize the importance of comprehensive, multi-dimensional evaluation to accurately characterize jailbreaking effectiveness and safety risks in large language models.

In many research areas, for example motion and gesture generation, objective measures alone do not provide an accurate impression of key stimulus traits such as perceived quality or appropriateness. The gold standard is instead to evaluate these aspects through user studies, especially subjective evaluations of video stimuli. Common evaluation paradigms either present individual stimuli to be scored on Likert-type scales, or ask users to compare and rate videos in a pairwise fashion. However, the time and resources required for such evaluations scale poorly as the number of conditions to be compared increases. Building on standards used for evaluating the quality of multimedia codecs, this paper instead introduces a framework for granular rating of multiple comparable videos in parallel. This methodology essentially analyses all condition pairs at once. Our contributions are 1) a proposed framework, called HEMVIP, for parallel and granular evaluation of multiple video stimuli and 2) a validation study confirming that results obtained using the tool are in close agreement with results of prior studies using conventional multiple pairwise comparisons.

Embodied agents benefit from using non-verbal behavior when communicating with humans. Despite several decades of non-verbal behavior-generation research, there is currently no well-developed benchmarking culture in the field. For example, most researchers do not compare their outcomes with previous work, and if they do, they often do so in their own way which frequently is incompatible with others. With the GENEA Workshop 2021, we aim to bring the community together to discuss key challenges and solutions, and find the most appropriate ways to move the field forward.

Co-speech gestures, gestures that accompany speech, play an important role in human communication. Automatic co-speech gesture generation is thus a key enabling technology for embodied conversational agents (ECAs), since humans expect ECAs to be capable of multi-modal communication. Research into gesture generation is rapidly gravitating towards data-driven methods. Unfortunately, individual research efforts in the field are difficult to compare: there are no established benchmarks, and each study tends to use its own dataset, motion visualisation, and evaluation methodology. To address this situation, we launched the GENEA Challenge, a gesture-generation challenge wherein participating teams built automatic gesture-generation systems on a common dataset, and the resulting systems were evaluated in parallel in a large, crowdsourced user study using the same motion-rendering pipeline. Since differences in evaluation outcomes between systems now are solely attributable to differences between the motion-generation methods, this enables benchmarking recent approaches against one another in order to get a better impression of the state of the art in the field. This paper reports on the purpose, design, results, and implications of our challenge.

We propose a new framework for gesture generation, aiming to allow data-driven approaches to produce more semantically rich gestures. Our approach first predicts whether to gesture, followed by a prediction of the gesture properties. Those properties are then used as conditioning for a modern probabilistic gesture-generation model capable of high-quality output. This empowers the approach to generate gestures that are both diverse and representational. Follow-ups and more information can be found on the project page: https://svito-zar.github.io/speech2properties2gestures/

During speech, people spontaneously gesticulate, which plays a key role in conveying information. Similarly, realistic co-speech gestures are crucial to enable natural and smooth interactions with social agents. Current end-to-end co-speech gesture generation systems use a single modality for representing speech: either audio or text. These systems are therefore confined to producing either acoustically-linked beat gestures or semantically-linked gesticulation (e.g., raising a hand when saying "high''): they cannot appropriately learn to generate both gesture types. We present a model designed to produce arbitrary beat and semantic gestures together. Our deep-learning based model takes both acoustic and semantic representations of speech as input, and generates gestures as a sequence of joint angle rotations as output. The resulting gestures can be applied to both virtual agents and humanoid robots. Subjective and objective evaluations confirm the success of our approach. The code and video are available at the project page svito-zar.github.io/gesticulator .

Conducting user studies is a crucial component in many scientific fields. While some studies require participants to be physically present, other studies can be conducted both physically (e.g. in-lab) and online (e.g. via crowdsourcing). Inviting participants to the lab can be a time-consuming and logistically difficult endeavor, not to mention that sometimes research groups might not be able to run in-lab experiments, because of, for example, a pandemic. Crowdsourcing platforms such as Amazon Mechanical Turk (AMT) or Prolific can therefore be a suitable alternative to run certain experiments, such as evaluating virtual agents. Although previous studies investigated the use of crowdsourcing platforms for running experiments, there is still uncertainty as to whether the results are reliable for perceptual studies. Here we replicate a previous experiment where participants evaluated a gesture generation model for virtual agents. The experiment is conducted across three participant pools - in-lab, Prolific, and AMT - having similar demographics across the in-lab participants and the Prolific platform. Our results show no difference between the three participant pools in regards to their evaluations of the gesture generation models and their reliability scores. The results indicate that online platforms can successfully be used for perceptual evaluations of this kind.

To enable more natural face-to-face interactions, conversational agents need to adapt their behavior to their interlocutors. One key aspect of this is generation of appropriate non-verbal behavior for the agent, for example facial gestures, here defined as facial expressions and head movements. Most existing gesture-generating systems do not utilize multi-modal cues from the interlocutor when synthesizing non-verbal behavior. Those that do, typically use deterministic methods that risk producing repetitive and non-vivid motions. In this paper, we introduce a probabilistic method to synthesize interlocutor-aware facial gestures - represented by highly expressive FLAME parameters - in dyadic conversations. Our contributions are: a) a method for feature extraction from multi-party video and speech recordings, resulting in a representation that allows for independent control and manipulation of expression and speech articulation in a 3D avatar; b) an extension to MoGlow, a recent motion-synthesis method based on normalizing flows, to also take multi-modal signals from the interlocutor as input and subsequently output interlocutor-aware facial gestures; and c) a subjective evaluation assessing the use and relative importance of the different modalities in the synthesized output. The results show that the model successfully leverages the input from the interlocutor to generate more appropriate behavior. Videos, data, and code are available at: https://jonepatr.github.io/lets_face_it/.

Feedback generation is an important component of humanhuman communication. Humans can choose to signal support, understanding, agreement or also sceptiscism by means of feedback tokens. Many studie ...

We propose a new way of eyes-free interaction for wearables. It is based on the user's proprioceptive sense, i.e., rather than seeing, hearing, or feeling an outside stimulus, users feel the pose of their own body. We have implemented a wearable device called Pose-IO that offers input and output based on proprioception. Users communicate with Pose-IO through the pose of their wrists. Users enter information by performing an input gesture by flexing their wrist, which the device senses using a 3-axis accelerometer. Users receive output from Pose-IO by find-ing their wrist posed in an output gesture, which Pose-IO actuates using electrical muscle stimulation. This mechanism allows users to interact with Pose-IO without visual or auditory senses, but through the proprioceptive sense alone. We developed three simple applications that demonstrate symmetric proprioceptive interaction, where input and output occur through the same limb, as well as asymmetric interaction, where input and output occur through different limbs. In a first user study, participants using a symmetric proprioceptive interface re-entered poses received from Pose-IO with an average accuracy of 5.8° despite the minimal bandwidth offered by the device. In a second, exploratory study, we investigated participants' emotional response to asymmetric proprioceptive interaction and the concept of the user's body serving as interface. Participants reported to enjoy the experience (4.6 out of 5).

We propose extending the affordance of objects by allowing them to communicate dynamic use, such as (1) motion (e.g., spray can shakes when touched), (2) multi-step processes (e.g., spray can sprays only after shaking), and (3) behaviors that change over time (e.g., empty spray can does not allow spraying anymore). Rather than enhancing objects directly, however, we implement this concept by enhancing the user. We call this affordance++. By stimulating the user's arms using electrical muscle stimulation, our prototype allows objects not only to make the user actuate them, but also perform required movements while merely approaching the object, such as not to touch objects that do not "want" to be touched. In our user study, affordance++ helped participants to successfully operate devices of poor natural affordance, such as a multi-functional slicer tool or a magnetic nail sweeper, and to stay away from cups filled with hot liquids.