Steering vectors are a lightweight method for controlling text properties by adding a learned bias to language model activations at inference time. So far, steering vectors have predominantly been evaluated in multiple-choice settings, while their effectiveness in free-form generation tasks remains understudied. Moving "Beyond Multiple Choice," we thoroughly evaluate the effectiveness of steering vectors in adaptively controlling topical focus, sentiment, toxicity, and readability in abstractive summaries of the NEWTS dataset. We find that steering effectively controls the targeted summary properties, but high steering strengths consistently degrade both intrinsic and extrinsic text quality. Compared to steering, prompting offers weaker control, while preserving text quality. Combining steering and prompting yields the strongest control over text properties and offers the most favorable efficacy-quality trade-off at moderate steering strengths. Our results underscore the practical trade-off between control strength and text quality preservation when applying steering vectors to free-form generation tasks.

As large language models (LLMs) start interacting with each other and generating an increasing amount of text online, it becomes crucial to better understand how information is transformed as it passes from one LLM to the next. While significant research has examined individual LLM behaviors, existing studies have largely overlooked the collective behaviors and information distortions arising from iterated LLM interactions. Small biases, negligible at the single output level, risk being amplified in iterated interactions, potentially leading the content to evolve towards attractor states. In a series of telephone game experiments, we apply a transmission chain design borrowed from the human cultural evolution literature: LLM agents iteratively receive, produce, and transmit texts from the previous to the next agent in the chain. By tracking the evolution of text toxicity, positivity, difficulty, and length across transmission chains, we uncover the existence of biases and attractors, and study their dependence on the initial text, the instructions, language model, and model size. For instance, we find that more open-ended instructions lead to stronger attraction effects compared to more constrained tasks. We also find that different text properties display different sensitivity to attraction effects, with toxicity leading to stronger attractors than length. These findings highlight the importance of accounting for multi-step transmission dynamics and represent a first step towards a more comprehensive understanding of LLM cultural dynamics.

You may have seen our previous posts on machine learning -- specifically, how to let your code learn from text and working with stop words, stemming, and spam. So today, we're going to build our machine learning-based spam filter, using the tools we walked through in those posts: tokenizer, stemmer, and naive bayes classifier. We are going to work with bluebird promise library here, so if you are not used to promises, please take a look at the bluebird API reference. Before we begin, it's important to have good training data. You can download some here -- we are interested in two.