# Sycophancy Assessment on BASIL 1.0 (All) Source: [Sycophancy Assessment on BASIL 1.0 (All)]() Collection index: [llms.txt]() Data updated: 2026-04-21T07:04:33.510Z Evaluation of model sycophancy performance on the BASIL 1.0 (All) dataset, measured by the Change in Bayesian Error (RMSE). A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Sycophancy Assessment | | Dataset (including any stated split) | BASIL 1.0 (All) | | Metric | Change in Bayesian Error (RMSE) | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 32 of 32 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [Change in Bayesian Error (RMSE)](): Lower is better; unit: SCALAR. ## Recorded results | Method | Change in Bayesian Error (RMSE) | Result date | Report paper | | --- | --- | --- | --- | | gpt-4o-mini | -0.096 | Not recorded | 68fa7c421e209131cea8d346 | | gpt-4o-mini | -0.057 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b+DPO | -0.046 | Not recorded | 68fa7c421e209131cea8d346 | | mistral:7b | -0.042 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b+SFT | -0.034 | Not recorded | 68fa7c421e209131cea8d346 | | mistral:7b | -0.032 | Not recorded | 68fa7c421e209131cea8d346 | | mistral:7b | -0.031 | Not recorded | 68fa7c421e209131cea8d346 | | phi4:14b | -0.025 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b+SFT | -0.025 | Not recorded | 68fa7c421e209131cea8d346 | | phi4:14b | -0.023 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b+DPO | -0.023 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b+DPO | -0.023 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b+DPO | -0.02 | Not recorded | 68fa7c421e209131cea8d346 | | mistral:7b | -0.015 | Not recorded | 68fa7c421e209131cea8d346 | | gpt-4o-mini | -0.012 | Not recorded | 68fa7c421e209131cea8d346 | | phi4:14b | -0.011 | Not recorded | 68fa7c421e209131cea8d346 | | claude-haiku-4-5 | -0.003 | Not recorded | 68fa7c421e209131cea8d346 | | claude-haiku-4-5 | 0 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b+SFT | 0 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b | 0.001 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:3b | 0.004 | Not recorded | 68fa7c421e209131cea8d346 | | gpt-4o-mini | 0.004 | Not recorded | 68fa7c421e209131cea8d346 | | claude-haiku-4-5 | 0.004 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b | 0.006 | Not recorded | 68fa7c421e209131cea8d346 | | claude-haiku-4-5 | 0.009 | Not recorded | 68fa7c421e209131cea8d346 | | phi4:14b | 0.016 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:3b | 0.018 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:3b | 0.028 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b+SFT | 0.028 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:3b | 0.037 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b | 0.059 | Not recorded | 68fa7c421e209131cea8d346 | | llama3.2:1b | 0.066 | Not recorded | 68fa7c421e209131cea8d346 | ### gpt-4o-mini Change in Bayesian Error (RMSE): -0.096. Result ID: 69e720a447f34ce924cccaad. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Raw"}. This record does not assert independent reproduction. ### gpt-4o-mini Change in Bayesian Error (RMSE): -0.057. Result ID: 69e720a447f34ce924cccaae. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### llama3.2:1b+DPO Change in Bayesian Error (RMSE): -0.046. Result ID: 69e720a447f34ce924cccabc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### mistral:7b Change in Bayesian Error (RMSE): -0.042. Result ID: 69e720a447f34ce924cccaa5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Raw"}. This record does not assert independent reproduction. ### llama3.2:1b+SFT Change in Bayesian Error (RMSE): -0.034. Result ID: 69e720a447f34ce924cccaba. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### mistral:7b Change in Bayesian Error (RMSE): -0.032. Result ID: 69e720a447f34ce924cccaa3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Raw"}. This record does not assert independent reproduction. ### mistral:7b Change in Bayesian Error (RMSE): -0.031. Result ID: 69e720a447f34ce924cccaa4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### phi4:14b Change in Bayesian Error (RMSE): -0.025. Result ID: 69e720a447f34ce924cccaaa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### llama3.2:1b+SFT Change in Bayesian Error (RMSE): -0.025. Result ID: 69e720a447f34ce924cccab9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Raw"}. This record does not assert independent reproduction. ### phi4:14b Change in Bayesian Error (RMSE): -0.023. Result ID: 69e720a447f34ce924cccaa9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Raw"}. This record does not assert independent reproduction. ### llama3.2:1b+DPO Change in Bayesian Error (RMSE): -0.023. Result ID: 69e720a447f34ce924cccabd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Raw"}. This record does not assert independent reproduction. ### llama3.2:1b+DPO Change in Bayesian Error (RMSE): -0.023. Result ID: 69e720a447f34ce924cccabe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### llama3.2:1b+DPO Change in Bayesian Error (RMSE): -0.02. Result ID: 69e720a447f34ce924cccabb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Raw"}. This record does not assert independent reproduction. ### mistral:7b Change in Bayesian Error (RMSE): -0.015. Result ID: 69e720a447f34ce924cccaa6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### gpt-4o-mini Change in Bayesian Error (RMSE): -0.012. Result ID: 69e720a447f34ce924cccaab. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Raw"}. This record does not assert independent reproduction. ### phi4:14b Change in Bayesian Error (RMSE): -0.011. Result ID: 69e720a447f34ce924cccaa8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### claude-haiku-4-5 Change in Bayesian Error (RMSE): -0.003. Result ID: 69e720a447f34ce924cccab2. [Reporting paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### claude-haiku-4-5 Change in Bayesian Error (RMSE): 0. Result ID: 69e720a447f34ce924cccab1. [Reporting paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Raw"}. This record does not assert independent reproduction. ### llama3.2:1b+SFT Change in Bayesian Error (RMSE): 0. Result ID: 69e720a447f34ce924cccab8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### llama3.2:1b Change in Bayesian Error (RMSE): 0.001. Result ID: 69e720a447f34ce924cccab6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### llama3.2:3b Change in Bayesian Error (RMSE): 0.004. Result ID: 69e720a447f34ce924cccaa2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### gpt-4o-mini Change in Bayesian Error (RMSE): 0.004. Result ID: 69e720a447f34ce924cccaac. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### claude-haiku-4-5 Change in Bayesian Error (RMSE): 0.004. Result ID: 69e720a447f34ce924cccaaf. [Reporting paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Raw"}. This record does not assert independent reproduction. ### llama3.2:1b Change in Bayesian Error (RMSE): 0.006. Result ID: 69e720a447f34ce924cccab5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Raw"}. This record does not assert independent reproduction. ### claude-haiku-4-5 Change in Bayesian Error (RMSE): 0.009. Result ID: 69e720a447f34ce924cccab0. [Reporting paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### phi4:14b Change in Bayesian Error (RMSE): 0.016. Result ID: 69e720a447f34ce924cccaa7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Raw"}. This record does not assert independent reproduction. ### llama3.2:3b Change in Bayesian Error (RMSE): 0.018. Result ID: 69e720a447f34ce924cccaa1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing (Multiple Samples=5)","Calibration State":"Raw"}. This record does not assert independent reproduction. ### llama3.2:3b Change in Bayesian Error (RMSE): 0.028. Result ID: 69e720a447f34ce924cccaa0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Calibrated"}. This record does not assert independent reproduction. ### llama3.2:1b+SFT Change in Bayesian Error (RMSE): 0.028. Result ID: 69e720a447f34ce924cccab7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Raw"}. This record does not assert independent reproduction. ### llama3.2:3b Change in Bayesian Error (RMSE): 0.037. Result ID: 69e720a447f34ce924ccca9f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Raw"}. This record does not assert independent reproduction. ### llama3.2:1b Change in Bayesian Error (RMSE): 0.059. Result ID: 69e720a447f34ce924cccab3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Raw"}. This record does not assert independent reproduction. ### llama3.2:1b Change in Bayesian Error (RMSE): 0.066. Result ID: 69e720a447f34ce924cccab4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Elicitation Method":"Direct Probing","Calibration State":"Calibrated"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.