Update README.md

README.md

@@ -18,7 +18,7 @@ language:
 ---
 
 <h1 align="center">
-  <img src="./assets/icons.png" alt="MARA Icon" width="45"  height="45"/> MARA AGENTS
+  <img src="icons.png" alt="MARA Icon" width="45"  height="45"/> MARA AGENTS
 </h1>
 <div style="display: flex; justify-content: center; gap: 10px;">
   <a href="https://github.com/IAAR-Shanghai/MARA">
@@ -32,7 +32,7 @@ language:
 
 **MARA** (Micro token-level Accept-Reject Alignment) simplifies the alignment process by breaking down sentence-level preference learning into fine-grained token-level binary classification. The MARA agent—a lightweight multi-layer perceptron (MLP)—operates as an alignment model that evaluates and classifies each candidate token as either *Accepted* or *Rejected* during LLM text generation.
 <figure>
-  <img src="./assets/mara_architecture.png" alt="mara_architecture" style="display: block; margin: 0 auto;" />
+  <img src="mara_architecture.png" alt="mara_architecture" style="display: block; margin: 0 auto;" />
   <figcaption style="text-align: center;">Architecture of MARA: The alignment model performs token selection through accept-reject decisions.</figcaption>
 </figure>
 
@@ -64,7 +64,7 @@ The source code and implementation details are open-sourced at [MARA](https://gi
 <table class="center">
     <tr>
         <td width=100% style="border: none">
-        <img src="assets/table1.png" style="width:50%; max-width:100%;">
+        <img src="table1.png" style="width:50%; max-width:100%;">
         <div style="text-align: left; margin-top: 8px;">Performance improvements of MARA across PKUSafeRLHF, BeaverTails, and HarmfulQA datasets. Each entry shows the percentage improvement in preference rate achieved by applying MARA compared to using the original LLM alone.</div>
         </td>
     </tr>
@@ -72,7 +72,7 @@ The source code and implementation details are open-sourced at [MARA](https://gi
 <table class="center">
     <tr>
         <td width=100% style="border: none">
-        <img src="assets/table3.png" style="width:50%; max-width:100%;">
+        <img src="table3.png" style="width:50%; max-width:100%;">
         <div style="text-align: left; margin-top: 8px;">Compatibility analysis of MARA, an alignment model trained with a LLM to be aggregate with other inference LLM. The value of each cell represents the percentage improvement in preference rate of our algorithm over the upstream model, i.e., inference model.</div>
         </td>
     </tr>
@@ -81,7 +81,7 @@ The source code and implementation details are open-sourced at [MARA](https://gi
 <table class="center">
     <tr>
         <td width=100% style="border: none">
-            <img src="assets/table2.png" style="width:100%">
+            <img src="table2.png" style="width:100%">
             <div style="text-align: left; margin-top: 8px;">Performance comparison of MARA against RLHF, DPO, and Aligner measured by percentage improvements of preference rate.</div>
         </td>
     </tr>