Goto

Collaborating Authors

 offensive statement


Intent-conditioned and Non-toxic Counterspeech Generation using Multi-Task Instruction Tuning with RLAIF

Hengle, Amey, Kumar, Aswini, Singh, Sahajpreet, Bandhakavi, Anil, Akhtar, Md Shad, Chakroborty, Tanmoy

arXiv.org Artificial Intelligence

Counterspeech, defined as a response to mitigate online hate speech, is increasingly used as a non-censorial solution. Addressing hate speech effectively involves dispelling the stereotypes, prejudices, and biases often subtly implied in brief, single-sentence statements or abuses. These implicit expressions challenge language models, especially in seq2seq tasks, as model performance typically excels with longer contexts. Our study introduces CoARL, a novel framework enhancing counterspeech generation by modeling the pragmatic implications underlying social biases in hateful statements. CoARL's first two phases involve sequential multi-instruction tuning, teaching the model to understand intents, reactions, and harms of offensive statements, and then learning task-specific low-rank adapter weights for generating intent-conditioned counterspeech. The final phase uses reinforcement learning to fine-tune outputs for effectiveness and non-toxicity. CoARL outperforms existing benchmarks in intent-conditioned counterspeech generation, showing an average improvement of 3 points in intent-conformity and 4 points in argument-quality metrics. Extensive human evaluation supports CoARL's efficacy in generating superior and more context-appropriate responses compared to existing systems, including prominent LLMs like ChatGPT.


Spoofing the Blenderbot

#artificialintelligence

Facebook became a known brand this century, but the iconic moniker was scrapped in favor of "Meta" in 2022. The latest from these lords of nomenclature is the Blenderbot 3, described in a blog post on ai.facebook.com The post, attributed to "Joelle Pineau, managing director of fundamental AI research at Meta," opens with a paragraph that begins by addressing "problematic or offensive language" and ends with a clunky evisceration of the English vernacular, to wit: "When we launched BlenderBot 3 a few days ago, we talked extensively about the promise and challenges that come with such a public demo, including the possibility that it could result in problematic or offensive language. While it is painful to see some of these offensive responses, public demos like this are important for building truly robust conversational AI systems and bridging the clear gap that exists today before such systems can be productionized." Frankenstein words like "productionized" should be edited out at this level, but never mind.

  Country:
  Industry: Information Technology > Services (0.35)

Meta's chatbot says the company 'exploits people'

#artificialintelligence

"Everyone who uses Blender Bot is required to acknowledge they understand it's for research and entertainment purposes only, that it can make untrue or offensive statements, and that they agree to not intentionally trigger the bot to make offensive statements," said a Meta spokesperson.


Meta's new AI chatbot can't stop bashing Facebook

The Guardian

If you're worried that artificial intelligence is getting too smart, talking to Meta's AI chatbot might make you feel better. Launched on Friday, BlenderBot is a prototype of Meta's conversational AI, which, according to Facebook's parent company, can converse on nearly any topic. On the demo website, members of the public are invited to chat with the tool and share feedback with developers. The results thus far, writers at Buzzfeed and Vice have pointed out, have been rather interesting. Asked about Mark Zuckerberg, the bot told BuzzFeed's Max Woolf that "he is a good businessman, but his business practices are not always ethical. It is funny that he has all this money and still wears the same clothes!"


5 Real-Life Examples of AI That Went Wrong

#artificialintelligence

The world is working on AI and day by day reaching a new chapter of it. Humanoid robots are going to be our coworkers, servers, and maybe even friends. This is what the motto is of robot makers. Although it hasn't been extremely successful yet, there are lots of AI that have been working in different areas and many expect that robots will be walking among us within the next 20–30 years. Among the many achievements, there were also some untoward incidents with AI that happened not so long ago.