United States

Large Language Models (LLMs) have shown remarkable progress, but their real-world application necessitates reliable calibration. This study conducts a comprehensive analysis of calibration degradation of LLMs across four dimensions: models, calibration metrics, tasks, and confidence extraction methods. Initial analysis showed that the relationship between alignment and calibration is not always a trade-off, but under stricter analysis conditions, we found the alignment process consistently harms calibration. This highlights the need for (1) a careful approach when measuring model confidences and calibration errors and (2) future research into algorithms that can help LLMs to achieve both instruction-following and calibration without sacrificing either.

EMNLP 2024

Does Alignment Tuning Really Break LLMs’ Internal Confidence?

calibration of large language models

alignment

## Welcome to EMNLP 2024! 
We are excited to welcome you to one of the most prominent conferences in the field of Natural Language Processing. This year, EMNLP 2024 is being held in a hybrid format,
offering both virtual and in-person participation in beautiful Miami. Due to a record-breaking number of submissions, we've expanded the total number of accepted papers to accommodate more cutting-edge research from around the globe.
### [Conference Handbook](https://drive.google.com/file/d/1WPROgxjLAC96AJL7Ugy0tEnYm7dkrbHt/view?usp=sharing)

You are required to register for this event. **Please register [here](https://2024.emnlp.org/registration/).** The EMNLP 2024 event page on Underline will be open to public one week prior to the event.

Please register!

EMNLP 2024 will take place in Miami, Florida from Nov 12th to Nov 16th, 2024, at the Hyatt Regency Miami Hote and on Underline for remote participants.

We present a comprehensive explainability dashboard designed for in-game chat toxicity. This dashboard integrates various existing explainable AI (XAI) techniques, including token importance analysis, model output visualization, and attribution to the training dataset. It also provides insights through the closest positive and negative examples, facilitating a deeper understanding and potential correction of the training data. Additionally, the dashboard includes word sense analysis—particularly useful for new moderators—and offers free-text explanations for both positive and negative predictions. This multi-faceted approach enhances the interpretability and transparency of toxicity detection models

ToxiSight: Insights Towards Detected Chat Toxicity

Semantic plausibility (e.g. knowing that “the actor won the award” is more likely than “the actor won the battle”) serves as an effective proxy for general world knowledge. Language models (LMs) capture vast amounts of world knowledge by learning distributional patterns in text, accessible via log probabilities they assign to plausible vs. implausible outputs. The new generation of instruction-tuned LMs can now also provide explicit estimates of plausibility via prompting. Here, we evaluate the effectiveness of log probabilities and basic prompting to measure semantic plausibility, both in single-sentence minimal pairs (Experiment 1) and short context- dependent scenarios (Experiment 2). We find that (i) in both base and instruction-tuned LMs, log probabilities offer a more reliable measure of semantic plausibility than direct zero-shot prompting, which yields inconsistent and often poor results; (ii) instruction-tuning generally does not alter the sensitivity of log probabilities to semantic plausibility (although sometimes decreases it); (iii) across models, context mostly modulates log probabilities in expected ways, as measured by three novel metrics of context-sensitive plausibility and their match to explicit human plausibility judgments. We conclude that, even in the era of prompt-based evaluations, log probabilities constitute a useful metric of semantic plausibility, both in base and instruction-tuned LMs.

Log Probabilities Are a Reliable Estimate of Semantic Plausibility in Base and Instruction-Tuned Language Models

Sparse autoencoders (SAEs) are an unsupervised method for learning a sparse decomposition of a neural network’s latent representations into seemingly interpretable features. Despite recent excitement about their potential, research applications outside of industry are limited by the high cost of training a comprehensive suite of SAEs. In this work, we introduce Gemma Scope, an open suite of JumpReLU SAEs trained on all layers and sub-layers of Gemma 2 2B and 9B and select layers of Gemma 2 27B base models. We primarily train SAEs on the Gemma 2 pre-trained models, but additionally release SAEs trained on instruction-tuned Gemma 2 9B for comparison. We evaluate the quality of each SAE on standard metrics and release these results. We hope that by releasing these SAE weights, we can help make more ambitious safety and interpretability research easier for the community. Weights and a tutorial can be found at https://huggingface.co/google/gemma-scope and an interactive demo can be found at https://neuronpedia.org/gemma-scope.

Gemma Scope: Open Sparse Autoencoders Everywhere All At Once on Gemma 2

Transformer-based language models have shown an excellent ability to effectively cap- ture and utilize contextual information. Al- though various analysis techniques have been used to quantify and trace the contribution of single contextual cues to a target task such as subject-verb agreement or coreference resolu- tion, scenarios in which multiple relevant cues are available in the context remain underex- plored. In this paper, we investigate how lan- guage models handle gender agreement when multiple gender cue words are present, each capable of independently disambiguating a tar- get gender pronoun. We analyze two widely used Transformer-based models: BERT, an encoder-based, and GPT-2, a decoder-based model. Our analysis employs two comple- mentary approaches: context mixing analysis, which tracks information flow within the model, and a variant of activation patching, which mea- sures the impact of cues on the model’s predic- tion. We find that BERT tends to prioritize the first cue in the context to form both the target word representations and the model’s predic- tion, while GPT-2 relies more on the final cue. Our findings reveal striking differences in how encoder-based and decoder-based models pri- oritize and use contextual information for their predictions.

How Language Models Prioritize Contextual Grammatical Cues?

Despite the widespread adoption of autoregressive language models, explainability evaluation research has predominantly focused on span infilling and masked language models. Evaluating the faithfulness of an explanation method—how accurately it explains the inner workings and decision-making of the model—is challenging because it is difficult to separate the model from its explanation. Most faithfulness evaluation techniques corrupt or remove input tokens deemed important by a particular attribution (feature importance) method and observe the resulting change in the model's output. However, for autoregressive language models, this approach creates out-of-distribution inputs due to their next-token prediction training objective. In this study, we propose a technique that leverages counterfactual generation to evaluate the faithfulness of attribution methods for autoregressive language models. Our technique generates fluent, in-distribution counterfactuals, making the evaluation protocol more reliable.

Counterfactuals As a Means for Evaluating Faithfulness of Attribution Methods in Autoregressive Language Models

Sparse AutoEncoders (SAEs) have gained popularity as a tool for enhancing the interpretability of Large Language Models (LLMs). However, training SAEs can be computationally intensive, especially as model complexity grows. In this study, the potential of transfer learning to accelerate SAEs training is explored by capitalizing on the shared representations found across adjacent layers of LLMs. Our experimental results demonstrate that fine-tuning SAEs using pre-trained models from nearby layers not only maintains but often improves the quality of learned representations, while significantly accelerating convergence. These findings indicate that the strategic reuse of pretrained SAEs is a promising approach, particularly in settings where computational resources are constrained.

Accelerating Sparse Autoencoder Training via Layer-Wise Transfer Learning in Large Language Models

Enriching Conceptual Knowledge in Language Models through Metaphorical Reference Explanation

Polish Coreference Corpus as an LLM Testbed: Evaluating Coreference Resolution within Instruction-Following Language Models by Instruction-Answer Alignment

Premium content

Does Alignment Tuning Really Break LLMs’ Internal Confidence?

Next from EMNLP 2024

ToxiSight: Insights Towards Detected Chat Toxicity

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES