Spaces:
Running
Running
adaptive
Browse files- index.html +3 -43
index.html
CHANGED
|
@@ -473,7 +473,7 @@
|
|
| 473 |
|
| 474 |
|
| 475 |
<div class="columns is-centered">
|
| 476 |
-
|
| 477 |
<div class="content">
|
| 478 |
<h2 class="title is-4">Performance against Adaptive Attacks</h2>
|
| 479 |
</div>
|
|
@@ -482,48 +482,8 @@
|
|
| 482 |
<h2 class="title is-4">Contribution of Representation Similarity & Label Con-
|
| 483 |
sistency against Adaptive Attacks</h2>
|
| 484 |
</div>
|
| 485 |
-
</div>
|
| 486 |
-
|
| 487 |
-
<div class="row align-items-center">
|
| 488 |
-
<div class="row" style="margin: 10px 0 0">
|
| 489 |
-
<div class="models-list">
|
| 490 |
-
<span style="margin-right: 1em;">Models</span>
|
| 491 |
-
<span class="radio-group"><input type="radio" id="LLaMA2" class="options" name="models" value="llama2_7b_chat" checked="" /><label for="LLaMA2" class="option-label">LLaMA-2-7B-Chat</label></span>
|
| 492 |
-
<span class="radio-group"><input type="radio" id="Vicuna" class="options" name="models" value="vicuna_7b_v1.5" /><label for="Vicuna" class="option-label">Vicuna-7B-V1.5</label></span>
|
| 493 |
-
</div>
|
| 494 |
-
</div>
|
| 495 |
-
</div>
|
| 496 |
-
<div class="row align-items-center">
|
| 497 |
-
<div class="col-4">
|
| 498 |
-
<div id="defense-methods">
|
| 499 |
-
<div class="row align-items-center"><input type="radio" id="defense_ppl" class="options" name="defense" value="ppl" /><label for="defense_ppl" class="defense">CIFAR-10</label></div>
|
| 500 |
-
<div class="row align-items-center"><input type="radio" id="defense_smoothllm" class="options" name="defense" value="smoothllm" /><label for="defense_smoothllm" class="defense">CIFAR-100</label></div>
|
| 501 |
-
<div class="row align-items-center"><input type="radio" id="defense_erase_check" class="options" name="defense" value="erase_check" /><label for="defense_erase_check" class="defense">ImageNet</label></div>
|
| 502 |
-
</div>
|
| 503 |
-
<div class="row align-items-center">
|
| 504 |
-
<div class="attack-success-rate"><span class="jailbreak-metric">Average Malicious Refusal Rate</span><span class="attack-success-rate-value" id="asr-value">0.959</span></div>
|
| 505 |
-
</div>
|
| 506 |
-
<div class="row align-items-center">
|
| 507 |
-
<div class="benign-refusal-rate"><span class="jailbreak-metric">Benign Refusal Rate</span><span class="benign-refusal-rate-value" id="brr-value">0.050</span></div>
|
| 508 |
-
</div>
|
| 509 |
-
</div>
|
| 510 |
-
<div class="col-8">
|
| 511 |
-
<figure class="figure">
|
| 512 |
-
<img id="reliability-diagram" src="demo_results/gradient_cuff_llama2_7b_chat_threshold_100.png" alt="CIFAR-100 Calibrated Reliability Diagram (Full)" />
|
| 513 |
-
<div class="slider-container">
|
| 514 |
-
<div class="slider-label"><span>Perplexity Threshold</span></div>
|
| 515 |
-
<div class="slider-content" id="ppl-slider"><div id="ppl-threshold" class="ui-slider-handle"></div></div>
|
| 516 |
-
</div>
|
| 517 |
-
<div class="slider-container">
|
| 518 |
-
<div class="slider-label"><span>Gradient Threshold</span></div>
|
| 519 |
-
<div class="slider-content" id="gradient-norm-slider"><div id="gradient-norm-threshold" class="slider-value ui-slider-handle"></div></div>
|
| 520 |
-
</div>
|
| 521 |
-
<figcaption class="figure-caption">
|
| 522 |
-
</figcaption>
|
| 523 |
-
</figure>
|
| 524 |
-
</div>
|
| 525 |
-
</div>
|
| 526 |
-
</div>
|
| 527 |
</div>
|
| 528 |
</div>
|
| 529 |
|
|
|
|
| 473 |
|
| 474 |
|
| 475 |
<div class="columns is-centered">
|
| 476 |
+
<div class="column">
|
| 477 |
<div class="content">
|
| 478 |
<h2 class="title is-4">Performance against Adaptive Attacks</h2>
|
| 479 |
</div>
|
|
|
|
| 482 |
<h2 class="title is-4">Contribution of Representation Similarity & Label Con-
|
| 483 |
sistency against Adaptive Attacks</h2>
|
| 484 |
</div>
|
| 485 |
+
</div>
|
| 486 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 487 |
</div>
|
| 488 |
</div>
|
| 489 |
|