- Rating
- 1
- Confidence
- 3
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
This work introduces a benchmark to analyse how LLMs perform premise correction under isolated and contextualised settings. The work finds LLMs usually can correct the false premises in isolated condition but not fail under contextualised condition, which is termed as correction suppression. They further conduct attention analysis in LLMs and claim that the contexts will divert the models' attention to the false premises so that the suppression is incurred. Finally, the work proposes two methods to mitigate the issue, and experiments show the effectiveness.
Critical strengths
-
This work introduces a meaningful "correction suppression" issue and constructs a benchmark to validate the existence of the issue. This may be helpful for LLM capacity improvement, such as hallucination reduction.
-
This work supplies an operationable pipeline to evaluate, analyse, and mitigate. This may provide inspiration for further studies on this issue.
-
Targeted to the issue, the work proposes two methods to mitigate it and experiments show its effectiveness.
Critical weaknesses
-
The reliable benchmark is the key basis to evaluate the issue, but the construction of the benchmark is not sufficiently introduced. There is much information that should be introduced in detail rather than simply listed out, such as raw materials collection, labelling processing, and verification processing.
-
The evaluation results are also not sufficient, and some experimental settings may be confused. For example, the models used in Tables 1 and 2 are not consistent, and I think the consistent models' employment will improve the reliability.
-
The analyses are also confused. For example, Figure 2 compares the hidden states of isolated and contextualised samples, but intuitively, it is worthy to compare between the suppressed and unsuppressed samples under contextualised setting.
-
The work claims the DPA method follows directly from their attention analysis. According to the experimental results, I find the performance of DPA does not significantly outperform the existing steering methods, which seems to suggest that the analysis does not motivate the underlying reasons for the issue.
-
To ensure reliability, it is worthy to use varying scaled models.
Questions for authors
-
How are the false premises and contexts generated, and how do you ensure the contexts align with the premises?
-
Why do you use the different models, such as in tables 1 and 2? What else meaningful insights evaluation results supply?
-
How the states and attention of suppressed and unsuppressed samples perform under contextualised setting? Are there any meaningful insights?
-
What the analysis inspires and why the directly followed DPA performs limited.
-
The over-correction results should be expanded, with at least more models to be added.